Search

In this paper, we present PCoTTA, an innovative, pioneering framework for Continual Test-Time Adaptation (CoTTA) in multi-task point cloud understanding, enhancing the model's transferability towards the continually changing target domain. We introduce a multi-task setting for PCoTTA, which is practical and realistic, handling multiple tasks within one unified model during the continual adaptation. Our PCoTTA involves three key components: automatic prototype mixture (APM), Gaussian Splatted feature shifting (GSFS), and contrastive prototype repulsion (CPR). Firstly, APM is designed to automatically mix the source prototypes with the learnable prototypes with a similarity balancing factor, avoiding catastrophic forgetting. Then, GSFS dynamically shifts the testing sample toward the source domain, mitigating error accumulation in an online manner. In addition, CPR is proposed to pull the nearest learnable prototype close to the testing feature and push it away from other prototypes, making each prototype distinguishable during the adaptation. Experimental comparisons lead to a new benchmark, demonstrating PCoTTA's superiority in boosting the model's transferability towards the continually changing target domain., Comment: Accepted to NeurIPS 2024

Speech-driven gesture generation aims at synthesizing a gesture sequence synchronized with the input speech signal. Previous methods leverage neural networks to directly map a compact audio representation to the gesture sequence, ignoring the semantic association of different modalities and failing to deal with salient gestures. In this paper, we propose a novel speech-driven gesture generation method by emphasizing the semantic consistency of salient posture. Specifically, we first learn a joint manifold space for the individual representation of audio and body pose to exploit the inherent semantic association between two modalities, and propose to enforce semantic consistency via a consistency loss. Furthermore, we emphasize the semantic consistency of salient postures by introducing a weakly-supervised detector to identify salient postures, and reweighting the consistency loss to focus more on learning the correspondence between salient postures and the high-level semantics of speech content. In addition, we propose to extract audio features dedicated to facial expression and body gesture separately, and design separate branches for face and body gesture synthesis. Extensive experimental results demonstrate the superiority of our method over the state-of-the-art approaches.

Masked point modeling methods have recently achieved great success in self-supervised learning for point cloud data. However, these methods are sensitive to rotations and often exhibit sharp performance drops when encountering rotational variations. In this paper, we propose a novel Rotation-Invariant Masked AutoEncoders (RI-MAE) to address two major challenges: 1) achieving rotation-invariant latent representations, and 2) facilitating self-supervised reconstruction in a rotation-invariant manner. For the first challenge, we introduce RI-Transformer, which features disentangled geometry content, rotation-invariant relative orientation and position embedding mechanisms for constructing rotation-invariant point cloud latent space. For the second challenge, a novel dual-branch student-teacher architecture is devised. It enables the self-supervised learning via the reconstruction of masked patches within the learned rotation-invariant latent space. Each branch is based on an RI-Transformer, and they are connected with an additional RI-Transformer predictor. The teacher encodes all point patches, while the student solely encodes unmasked ones. Finally, the predictor predicts the latent features of the masked patches using the output latent embeddings from the student, supervised by the outputs from the teacher. Extensive experiments demonstrate that our method is robust to rotations, achieving the state-of-the-art performance on various downstream tasks.

Extreme Multimodal Summarization with Multimodal Output (XMSMO) becomes an attractive summarization approach by integrating various types of information to create extremely concise yet informative summaries for individual modalities. Existing methods overlook the issue that multimodal data often contains more topic irrelevant information, which can mislead the model into producing inaccurate summaries especially for extremely short ones. In this paper, we propose SITransformer, a Shared Information-guided Transformer for extreme multimodal summarization. It has a shared information guided pipeline which involves a cross-modal shared information extractor and a cross-modal interaction module. The extractor formulates semantically shared salient information from different modalities by devising a novel filtering process consisting of a differentiable top-k selector and a shared-information guided gating unit. As a result, the common, salient, and relevant contents across modalities are identified. Next, a transformer with cross-modal attentions is developed for intra- and inter-modality learning with the shared information guidance to produce the extreme summary. Comprehensive experiments demonstrate that SITransformer significantly enhances the summarization quality for both video and text summaries for XMSMO. Our code will be publicly available at https://github.com/SichengLeoLiu/MMAsia24-XMSMO., Comment: 8 pages, 5 figures, submitted to ACM Multimedia Asia 2024

Domain Generalization (DG) has been recently explored to improve the generalizability of point cloud classification (PCC) models toward unseen domains. However, they often suffer from limited receptive fields or quadratic complexity due to the use of convolution neural networks or vision Transformers. In this paper, we present the first work that studies the generalizability of state space models (SSMs) in DG PCC and find that directly applying SSMs into DG PCC will encounter several challenges: the inherent topology of the point cloud tends to be disrupted and leads to noise accumulation during the serialization stage. Besides, the lack of designs in domain-agnostic feature learning and data scanning will introduce unanticipated domain-specific information into the 3D sequence data. To this end, we propose a novel framework, PointDGMamba, that excels in strong generalizability toward unseen domains and has the advantages of global receptive fields and efficient linear complexity. PointDGMamba consists of three innovative components: Masked Sequence Denoising (MSD), Sequence-wise Cross-domain Feature Aggregation (SCFA), and Dual-level Domain Scanning (DDS). In particular, MSD selectively masks out the noised point tokens of the point cloud sequences, SCFA introduces cross-domain but same-class point cloud features to encourage the model to learn how to extract more generalized features. DDS includes intra-domain scanning and cross-domain scanning to facilitate information exchange between features. In addition, we propose a new and more challenging benchmark PointDG-3to1 for multi-domain generalization. Extensive experiments demonstrate the effectiveness and state-of-the-art performance of our presented PointDGMamba.

Recent point cloud understanding research suffers from performance drops on unseen data, due to the distribution shifts across different domains. While recent studies use Domain Generalization (DG) techniques to mitigate this by learning domain-invariant features, most are designed for a single task and neglect the potential of testing data. Despite In-Context Learning (ICL) showcasing multi-task learning capability, it usually relies on high-quality context-rich data and considers a single dataset, and has rarely been studied in point cloud understanding. In this paper, we introduce a novel, practical, multi-domain multi-task setting, handling multiple domains and multiple tasks within one unified model for domain generalized point cloud understanding. To this end, we propose Domain Generalized Point-In-Context Learning (DG-PIC) that boosts the generalizability across various tasks and domains at testing time. In particular, we develop dual-level source prototype estimation that considers both global-level shape contextual and local-level geometrical structures for representing source domains and a dual-level test-time feature shifting mechanism that leverages both macro-level domain semantic information and micro-level patch positional relationships to pull the target data closer to the source ones during the testing. Our DG-PIC does not require any model updates during the testing and can handle unseen domains and multiple tasks, \textit{i.e.,} point cloud reconstruction, denoising, and registration, within one unified model. We also introduce a benchmark for this new setting. Comprehensive experiments demonstrate that DG-PIC outperforms state-of-the-art techniques significantly., Comment: Accepted to ECCV 2024

This paper introduces a new approach for generating globally consistent normals for point clouds sampled from manifold surfaces. Given that the generalized winding number (GWN) field generated by a point cloud with globally consistent normals is a solution to a PDE with jump boundary conditions and possesses harmonic properties, and the Dirichlet energy of the GWN field can be defined as an integral over the boundary surface, we formulate a boundary energy derived from the Dirichlet energy of the GWN. Taking as input a point cloud with randomly oriented normals, we optimize this energy to restore the global harmonicity of the GWN field, thereby recovering the globally consistent normals. Experiments show that our method outperforms state-of-the-art approaches, exhibiting enhanced robustness to noise, outliers, complex topologies, and thin structures. Our code can be found at \url{https://github.com/liuweizhou319/BIM}., Comment: accepted in siggraph2024

Adversarial attacks on Face Recognition (FR) systems have proven highly effective in compromising pure FR models, yet adversarial examples may be ineffective to the complete FR systems as Face Anti-Spoofing (FAS) models are often incorporated and can detect a significant number of them. To address this under-explored and essential problem, we propose a novel setting of adversarially attacking both FR and FAS models simultaneously, aiming to enhance the practicability of adversarial attacks on FR systems. In particular, we introduce a new attack method, namely Style-aligned Distribution Biasing (SDB), to improve the capacity of black-box attacks on both FR and FAS models. Specifically, our SDB framework consists of three key components. Firstly, to enhance the transferability of FAS models, we design a Distribution-aware Score Biasing module to optimize adversarial face examples away from the distribution of spoof images utilizing scores. Secondly, to mitigate the substantial style differences between live images and adversarial examples initialized with spoof images, we introduce an Instance Style Alignment module that aligns the style of adversarial examples with live images. In addition, to alleviate the conflicts between the gradients of FR and FAS models, we propose a Gradient Consistency Maintenance module to minimize disparities between the gradients using Hessian approximation. Extensive experiments showcase the superiority of our proposed attack method to state-of-the-art adversarial attacks.

Point cloud filtering is a fundamental 3D vision task, which aims to remove noise while recovering the underlying clean surfaces. State-of-the-art methods remove noise by moving noisy points along stochastic trajectories to the clean surfaces. These methods often require regularization within the training objective and/or during post-processing, to ensure fidelity. In this paper, we introduce StraightPCF, a new deep learning based method for point cloud filtering. It works by moving noisy points along straight paths, thus reducing discretization errors while ensuring faster convergence to the clean surfaces. We model noisy patches as intermediate states between high noise patch variants and their clean counterparts, and design the VelocityModule to infer a constant flow velocity from the former to the latter. This constant flow leads to straight filtering trajectories. In addition, we introduce a DistanceModule that scales the straight trajectory using an estimated distance scalar to attain convergence near the clean surface. Our network is lightweight and only has $\sim530K$ parameters, being 17% of IterativePFN (a most recent point cloud filtering network). Extensive experiments on both synthetic and real-world data show our method achieves state-of-the-art results. Our method also demonstrates nice distributions of filtered points without the need for regularization. The implementation code can be found at: https://github.com/ddsediri/StraightPCF., Comment: This paper has been accepted to the IEEE/CVF CVPR Conference, 2024

Domain generalization~(DG) aims at solving distribution shift problems in various scenes. Existing approaches are based on Convolution Neural Networks (CNNs) or Vision Transformers (ViTs), which suffer from limited receptive fields or quadratic complexities issues. Mamba, as an emerging state space model (SSM), possesses superior linear complexity and global receptive fields. Despite this, it can hardly be applied to DG to address distribution shifts, due to the hidden state issues and inappropriate scan mechanisms. In this paper, we propose a novel framework for DG, named DGMamba, that excels in strong generalizability toward unseen domains and meanwhile has the advantages of global receptive fields, and efficient linear complexity. Our DGMamba compromises two core components: Hidden State Suppressing~(HSS) and Semantic-aware Patch refining~(SPR). In particular, HSS is introduced to mitigate the influence of hidden states associated with domain-specific features during output prediction. SPR strives to encourage the model to concentrate more on objects rather than context, consisting of two designs: Prior-Free Scanning~(PFS), and Domain Context Interchange~(DCI). Concretely, PFS aims to shuffle the non-semantic patches within images, creating more flexible and effective sequences from images, and DCI is designed to regularize Mamba with the combination of mismatched non-semantic and semantic information by fusing patches among domains. Extensive experiments on five commonly used DG benchmarks demonstrate that the proposed DGMamba achieves remarkably superior results to state-of-the-art models. The code will be made publicly available at https://github.com/longshaocong/DGMamba., Comment: Accepted to ACM MM 2024

Face Anti-Spoofing (FAS) is pivotal in safeguarding facial recognition systems against presentation attacks. While domain generalization (DG) methods have been developed to enhance FAS performance, they predominantly focus on learning domain-invariant features during training, which may not guarantee generalizability to unseen data that differs largely from the source distributions. Our insight is that testing data can serve as a valuable resource to enhance the generalizability beyond mere evaluation for DG FAS. In this paper, we introduce a novel Test-Time Domain Generalization (TTDG) framework for FAS, which leverages the testing data to boost the model's generalizability. Our method, consisting of Test-Time Style Projection (TTSP) and Diverse Style Shifts Simulation (DSSS), effectively projects the unseen data to the seen domain space. In particular, we first introduce the innovative TTSP to project the styles of the arbitrarily unseen samples of the testing distribution to the known source space of the training distributions. We then design the efficient DSSS to synthesize diverse style shifts via learnable style bases with two specifically designed losses in a hyperspherical feature space. Our method eliminates the need for model updates at the test time and can be seamlessly integrated into not only the CNN but also ViT backbones. Comprehensive experiments on widely used cross-domain FAS benchmarks demonstrate our method's state-of-the-art performance and effectiveness., Comment: Accepted to IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

Segment anything model (SAM) has demonstrated excellent generalizability in common vision scenarios, yet falling short of the ability to understand specialized data. Recently, several methods have combined parameter-efficient techniques with task-specific designs to fine-tune SAM on particular tasks. However, these methods heavily rely on handcraft, complicated, and task-specific designs, and pre/post-processing to achieve acceptable performances on downstream tasks. As a result, this severely restricts generalizability to other downstream tasks. To address this issue, we present a simple and unified framework, namely SU-SAM, that can easily and efficiently fine-tune the SAM model with parameter-efficient techniques while maintaining excellent generalizability toward various downstream tasks. SU-SAM does not require any task-specific designs and aims to improve the adaptability of SAM-like models significantly toward underperformed scenes. Concretely, we abstract parameter-efficient modules of different methods into basic design elements in our framework. Besides, we propose four variants of SU-SAM, i.e., series, parallel, mixed, and LoRA structures. Comprehensive experiments on nine datasets and six downstream tasks to verify the effectiveness of SU-SAM, including medical image segmentation, camouflage object detection, salient object segmentation, surface defect segmentation, complex object shapes, and shadow masking. Our experimental results demonstrate that SU-SAM achieves competitive or superior accuracy compared to state-of-the-art methods. Furthermore, we provide in-depth analyses highlighting the effectiveness of different parameter-efficient designs within SU-SAM. In addition, we propose a generalized model and benchmark, showcasing SU-SAM's generalizability across all diverse datasets simultaneously.

Face Recognition (FR) systems can be easily deceived by adversarial examples that manipulate benign face images through imperceptible perturbations. Adversarial attacks on FR encompass two types: impersonation (targeted) attacks and dodging (untargeted) attacks. Previous methods often achieve a successful impersonation attack on FR, however, it does not necessarily guarantee a successful dodging attack on FR in the black-box setting. In this paper, our key insight is that the generation of adversarial examples should perform both impersonation and dodging attacks simultaneously. To this end, we propose a novel attack method termed as Adversarial Pruning (Adv-Pruning), to fine-tune existing adversarial examples to enhance their dodging capabilities while preserving their impersonation capabilities. Adv-Pruning consists of Priming, Pruning, and Restoration stages. Concretely, we propose Adversarial Priority Quantification to measure the region-wise priority of original adversarial perturbations, identifying and releasing those with minimal impact on absolute model output variances. Then, Biased Gradient Adaptation is presented to adapt the adversarial examples to traverse the decision boundaries of both the attacker and victim by adding perturbations favoring dodging attacks on the vacated regions, preserving the prioritized features of the original perturbations while boosting dodging performance. As a result, we can maintain the impersonation capabilities of original adversarial examples while effectively enhancing dodging capabilities. Comprehensive experiments demonstrate the superiority of our method compared with state-of-the-art adversarial attack methods., Comment: Accepted to ACM MM 2024

Recent works attempt to extend Graph Convolution Networks (GCNs) to point clouds for classification and segmentation tasks. These works tend to sample and group points to create smaller point sets locally and mainly focus on extracting local features through GCNs, while ignoring the relationship between point sets. In this paper, we propose the Dynamic Hop Graph Convolution Network (DHGCN) for explicitly learning the contextual relationships between the voxelized point parts, which are treated as graph nodes. Motivated by the intuition that the contextual information between point parts lies in the pairwise adjacent relationship, which can be depicted by the hop distance of the graph quantitatively, we devise a novel self-supervised part-level hop distance reconstruction task and design a novel loss function accordingly to facilitate training. In addition, we propose the Hop Graph Attention (HGA), which takes the learned hop distance as input for producing attention weights to allow edge features to contribute distinctively in aggregation. Eventually, the proposed DHGCN is a plug-and-play module that is compatible with point-based backbone networks. Comprehensive experiments on different backbones and tasks demonstrate that our self-supervised method achieves state-of-the-art performance. Our source code is available at: https://github.com/Jinec98/DHGCN., Comment: Accepted to AAAI 2024

In this paper, we address the challenge of image resolution variation for the Segment Anything Model (SAM). SAM, known for its zero-shot generalizability, exhibits a performance degradation when faced with datasets with varying image sizes. Previous approaches tend to resize the image to a fixed size or adopt structure modifications, hindering the preservation of SAM's rich prior knowledge. Besides, such task-specific tuning necessitates a complete retraining of the model, which is cost-expensive and unacceptable for deployment in the downstream tasks. In this paper, we reformulate this issue as a length extrapolation problem, where token sequence length varies while maintaining a consistent patch size for images of different sizes. To this end, we propose Scalable Bias-Mode Attention Mask (BA-SAM) to enhance SAM's adaptability to varying image resolutions while eliminating the need for structure modifications. Firstly, we introduce a new scaling factor to ensure consistent magnitude in the attention layer's dot product values when the token sequence length changes. Secondly, we present a bias-mode attention mask that allows each token to prioritize neighboring information, mitigating the impact of untrained distant information. Our BA-SAM demonstrates efficacy in two scenarios: zero-shot and fine-tuning. Extensive evaluation on diverse datasets, including DIS5K, DUTS, ISIC, COD10K, and COCO, reveals its ability to significantly mitigate performance degradation in the zero-shot setting and achieve state-of-the-art performance with minimal fine-tuning. Furthermore, we propose a generalized model and benchmark, showcasing BA-SAM's generalizability across all four datasets simultaneously. Code is available at https://github.com/zongzi13545329/BA-SAM, Comment: Accepted to IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

This paper explores a novel setting called Generalized Category Discovery in Semantic Segmentation (GCDSS), aiming to segment unlabeled images given prior knowledge from a labeled set of base classes. The unlabeled images contain pixels of the base class or novel class. In contrast to Novel Category Discovery in Semantic Segmentation (NCDSS), there is no prerequisite for prior knowledge mandating the existence of at least one novel class in each unlabeled image. Besides, we broaden the segmentation scope beyond foreground objects to include the entire image. Existing NCDSS methods rely on the aforementioned priors, making them challenging to truly apply in real-world situations. We propose a straightforward yet effective framework that reinterprets the GCDSS challenge as a task of mask classification. Additionally, we construct a baseline method and introduce the Neighborhood Relations-Guided Mask Clustering Algorithm (NeRG-MaskCA) for mask categorization to address the fragmentation in semantic representation. A benchmark dataset, Cityscapes-GCD, derived from the Cityscapes dataset, is established to evaluate the GCDSS framework. Our method demonstrates the feasibility of the GCDSS problem and the potential for discovering and segmenting novel object classes in unlabeled images. We employ the generated pseudo-labels from our approach as ground truth to supervise the training of other models, thereby enabling them with the ability to segment novel classes. It paves the way for further research in generalized category discovery, broadening the horizons of semantic segmentation and its applications. For details, please visit https://github.com/JethroPeng/GCDSS

Existing normal estimation methods for point clouds are often less robust to severe noise and complex geometric structures. Also, they usually ignore the contributions of different neighbouring points during normal estimation, which leads to less accurate results. In this paper, we introduce a weighted normal estimation method for 3D point cloud data. We innovate in two key points: 1) we develop a novel weighted normal regression technique that predicts point-wise weights from local point patches and use them for robust, feature-preserving normal regression; 2) we propose to conduct contrastive learning between point patches and the corresponding ground-truth normals of the patches' central points as a pre-training process to facilitate normal regression. Comprehensive experiments demonstrate that our method can robustly handle noisy and complex point clouds, achieving state-of-the-art performance on both synthetic and real-world datasets., Comment: Accepted by ICME 2023

Recall one time when we were in an unfamiliar mall. We might mistakenly think that there exists or does not exist a piece of glass in front of us. Such mistakes will remind us to walk more safely and freely at the same or a similar place next time. To absorb the human mistake correction wisdom, we propose a novel glass segmentation network to detect transparent glass, dubbed GlassSegNet. Motivated by this human behavior, GlassSegNet utilizes two key stages: the identification stage (IS) and the correction stage (CS). The IS is designed to simulate the detection procedure of human recognition for identifying transparent glass by global context and edge information. The CS then progressively refines the coarse prediction by correcting mistake regions based on gained experience. Extensive experiments show clear improvements of our GlassSegNet over thirty-four state-of-the-art methods on three benchmark datasets.

Face anti-spoofing (FAS) based on domain generalization (DG) has been recently studied to improve the generalization on unseen scenarios. Previous methods typically rely on domain labels to align the distribution of each domain for learning domain-invariant representations. However, artificial domain labels are coarse-grained and subjective, which cannot reflect real domain distributions accurately. Besides, such domain-aware methods focus on domain-level alignment, which is not fine-grained enough to ensure that learned representations are insensitive to domain styles. To address these issues, we propose a novel perspective for DG FAS that aligns features on the instance level without the need for domain labels. Specifically, Instance-Aware Domain Generalization framework is proposed to learn the generalizable feature by weakening the features' sensitivity to instance-specific styles. Concretely, we propose Asymmetric Instance Adaptive Whitening to adaptively eliminate the style-sensitive feature correlation, boosting the generalization. Moreover, Dynamic Kernel Generator and Categorical Style Assembly are proposed to first extract the instance-specific features and then generate the style-diversified features with large style shifts, respectively, further facilitating the learning of style-insensitive features. Extensive experiments and analysis demonstrate the superiority of our method over state-of-the-art competitors. Code will be publicly available at https://github.com/qianyuzqy/IADG., Comment: Accepted to IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023

The quality of point clouds is often limited by noise introduced during their capture process. Consequently, a fundamental 3D vision task is the removal of noise, known as point cloud filtering or denoising. State-of-the-art learning based methods focus on training neural networks to infer filtered displacements and directly shift noisy points onto the underlying clean surfaces. In high noise conditions, they iterate the filtering process. However, this iterative filtering is only done at test time and is less effective at ensuring points converge quickly onto the clean surfaces. We propose IterativePFN (iterative point cloud filtering network), which consists of multiple IterationModules that model the true iterative filtering process internally, within a single network. We train our IterativePFN network using a novel loss function that utilizes an adaptive ground truth target at each iteration to capture the relationship between intermediate filtering results during training. This ensures that the filtered results converge faster to the clean surfaces. Our method is able to obtain better performance compared to state-of-the-art methods. The source code can be found at: https://github.com/ddsediri/IterativePFN., Comment: This paper has been accepted to the IEEE/CVF CVPR Conference, 2023

Semi-supervised learning (SSL) has attracted enormous attention due to its vast potential of mitigating the dependence on large labeled datasets. The latest methods (e.g., FixMatch) use a combination of consistency regularization and pseudo-labeling to achieve remarkable successes. However, these methods all suffer from the waste of complicated examples since all pseudo-labels have to be selected by a high threshold to filter out noisy ones. Hence, the examples with ambiguous predictions will not contribute to the training phase. For better leveraging all unlabeled examples, we propose two novel techniques: Entropy Meaning Loss (EML) and Adaptive Negative Learning (ANL). EML incorporates the prediction distribution of non-target classes into the optimization objective to avoid competition with target class, and thus generating more high-confidence predictions for selecting pseudo-label. ANL introduces the additional negative pseudo-label for all unlabeled data to leverage low-confidence examples. It adaptively allocates this label by dynamically evaluating the top-k performance of the model. EML and ANL do not introduce any additional parameter and hyperparameter. We integrate these techniques with FixMatch, and develop a simple yet powerful framework called FullMatch. Extensive experiments on several common SSL benchmarks (CIFAR-10/100, SVHN, STL-10 and ImageNet) demonstrate that FullMatch exceeds FixMatch by a large margin. Integrated with FlexMatch (an advanced FixMatch-based framework), we achieve state-of-the-art performance. Source code is at https://github.com/megvii-research/FullMatch., Comment: Accepted by CVPR 2023

For pursuing accurate skeleton-based action recognition, most prior methods use the strategy of combining Graph Convolution Networks (GCNs) with attention-based methods in a serial way. However, they regard the human skeleton as a complete graph, resulting in less variations between different actions (e.g., the connection between the elbow and head in action ``clapping hands''). For this, we propose a novel Contrastive GCN-Transformer Network (ConGT) which fuses the spatial and temporal modules in a parallel way. The ConGT involves two parallel streams: Spatial-Temporal Graph Convolution stream (STG) and Spatial-Temporal Transformer stream (STT). The STG is designed to obtain action representations maintaining the natural topology structure of the human skeleton. The STT is devised to acquire action representations containing the global relationships among joints. Since the action representations produced from these two streams contain different characteristics, and each of them knows little information of the other, we introduce the contrastive learning paradigm to guide their output representations of the same sample to be as close as possible in a self-supervised manner. Through the contrastive learning, they can learn information from each other to enrich the action features by maximizing the mutual information between the two types of action representations. To further improve action recognition accuracy, we introduce the Cyclical Focal Loss (CFL) which can focus on confident training samples in early training epochs, with an increasing focus on hard samples during the middle epochs. We conduct experiments on three benchmark datasets, which demonstrate that our model achieves state-of-the-art performance in action recognition., Comment: 14 pages, 9 figures

Deep learning has enabled realistic face manipulation (i.e., deepfake), which poses significant concerns over the integrity of the media in circulation. Most existing deep learning techniques for deepfake detection can achieve promising performance in the intra-dataset evaluation setting (i.e., training and testing on the same dataset), but are unable to perform satisfactorily in the inter-dataset evaluation setting (i.e., training on one dataset and testing on another). Most of the previous methods use the backbone network to extract global features for making predictions and only employ binary supervision (i.e., indicating whether the training instances are fake or authentic) to train the network. Classification merely based on the learning of global features leads often leads to weak generalizability to unseen manipulation methods. In addition, the reconstruction task can improve the learned representations. In this paper, we introduce a novel approach for deepfake detection, which considers the reconstruction and classification tasks simultaneously to address these problems. This method shares the information learned by one task with the other, which focuses on a different aspect other existing works rarely consider and hence boosts the overall performance. In particular, we design a two-branch Convolutional AutoEncoder (CAE), in which the Convolutional Encoder used to compress the feature map into the latent representation is shared by both branches. Then the latent representation of the input data is fed to a simple classifier and the unsupervised reconstruction component simultaneously. Our network is trained end-to-end. Experiments demonstrate that our method achieves state-of-the-art performance on three commonly-used datasets, particularly in the cross-dataset evaluation setting., Comment: This paper has been submitted to Pattern Recognition Letter for review

Spectral graph convolutional neural networks (GCNNs) have been producing encouraging results in graph classification tasks. However, most spectral GCNNs utilize fixed graphs when aggregating node features, while omitting edge feature learning and failing to get an optimal graph structure. Moreover, many existing graph datasets do not provide initialized edge features, further restraining the ability of learning edge features via spectral GCNNs. In this paper, we try to address this issue by designing an edge feature scheme and an add-on layer between every two stacked graph convolution layers in GCNN. Both are lightweight while effective in filling the gap between edge feature learning and performance enhancement of graph classification. The edge feature scheme makes edge features adapt to node representations at different graph convolution layers. The add-on layers help adjust the edge features to an optimal graph structure. To test the effectiveness of our method, we take Euclidean positions as initial node features and extract graphs with semantic information from point cloud objects. The node features of our extracted graphs are more scalable for edge feature learning than most existing graph datasets (in one-hot encoded label format). Three new graph datasets are constructed based on ModelNet40, ModelNet10 and ShapeNet Part datasets. Experimental results show that our method outperforms state-of-the-art graph classification methods on the new datasets by reaching 96.56% overall accuracy on Graph-ModelNet40, 98.79% on Graph-ModelNet10 and 97.91% on Graph-ShapeNet Part. The constructed graph datasets will be released to the community.

How will you repair a physical object with large missings? You may first recover its global yet coarse shape and stepwise increase its local details. We are motivated to imitate the above physical repair procedure to address the point cloud completion task. We propose a novel stepwise point cloud completion network (SPCNet) for various 3D models with large missings. SPCNet has a hierarchical bottom-to-up network architecture. It fulfills shape completion in an iterative manner, which 1) first infers the global feature of the coarse result; 2) then infers the local feature with the aid of global feature; and 3) finally infers the detailed result with the help of local feature and coarse result. Beyond the wisdom of simulating the physical repair, we newly design a cycle loss %based training strategy to enhance the generalization and robustness of SPCNet. Extensive experiments clearly show the superiority of our SPCNet over the state-of-the-art methods on 3D point clouds with large missings.

In-place gesture-based virtual locomotion techniques enable users to control their viewpoint and intuitively move in the 3D virtual environment. A key research problem is to accurately and quickly recognize in-place gestures, since they can trigger specific movements of virtual viewpoints and enhance user experience. However, to achieve real-time experience, only short-term sensor sequence data (up to about 300ms, 6 to 10 frames) can be taken as input, which actually affects the classification performance due to limited spatio-temporal information. In this paper, we propose a novel long-term memory augmented network for in-place gestures classification. It takes as input both short-term gesture sequence samples and their corresponding long-term sequence samples that provide extra relevant spatio-temporal information in the training phase. We store long-term sequence features with an external memory queue. In addition, we design a memory augmented loss to help cluster features of the same class and push apart features from different classes, thus enabling our memory queue to memorize more relevant long-term sequence features. In the inference phase, we input only short-term sequence samples to recall the stored features accordingly, and fuse them together to predict the gesture class. We create a large-scale in-place gestures dataset from 25 participants with 11 gestures. Our method achieves a promising accuracy of 95.1% with a latency of 192ms, and an accuracy of 97.3% with a latency of 312ms, and is demonstrated to be superior to recent in-place gesture classification techniques. User study also validates our approach. Our source code and dataset will be made available to the community., Comment: This paper is accepted to IEEE ISMAR2022

6D pose estimation of rigid objects from RGB-D images is crucial for object grasping and manipulation in robotics. Although RGB channels and the depth (D) channel are often complementary, providing respectively the appearance and geometry information, it is still non-trivial how to fully benefit from the two cross-modal data. From the simple yet new observation, when an object rotates, its semantic label is invariant to the pose while its keypoint offset direction is variant to the pose. To this end, we present SO(3)-Pose, a new representation learning network to explore SO(3)-equivariant and SO(3)-invariant features from the depth channel for pose estimation. The SO(3)-invariant features facilitate to learn more distinctive representations for segmenting objects with similar appearance from RGB channels. The SO(3)-equivariant features communicate with RGB features to deduce the (missed) geometry for detecting keypoints of an object with the reflective surface from the depth channel. Unlike most of existing pose estimation methods, our SO(3)-Pose not only implements the information communication between the RGB and depth channels, but also naturally absorbs the SO(3)-equivariance geometry knowledge from depth images, leading to better appearance and geometry representation learning. Comprehensive experiments show that our method achieves the state-of-the-art performance on three benchmarks.

Point cloud filtering and normal estimation are two fundamental research problems in the 3D field. Existing methods usually perform normal estimation and filtering separately and often show sensitivity to noise and/or inability to preserve sharp geometric features such as corners and edges. In this paper, we propose a novel deep learning method to jointly estimate normals and filter point clouds. We first introduce a 3D patch based contrastive learning framework, with noise corruption as an augmentation, to train a feature encoder capable of generating faithful representations of point cloud patches while remaining robust to noise. These representations are consumed by a simple regression network and supervised by a novel joint loss, simultaneously estimating point normals and displacements that are used to filter the patch centers. Experimental results show that our method well supports the two tasks simultaneously and preserves sharp features and fine details. It generally outperforms state-of-the-art techniques on both tasks. Our source code is available at https://github.com/ddsediri/CLJNEPCF., Comment: This paper has been accepted to the IEEE TVCG journal. Our source code is available at https://github.com/ddsediri/CLJNEPCF

Transformer-based Self-supervised Representation Learning methods learn generic features from unlabeled datasets for providing useful network initialization parameters for downstream tasks. Recently, self-supervised learning based upon masking local surface patches for 3D point cloud data has been under-explored. In this paper, we propose masked Autoencoders in 3D point cloud representation learning (abbreviated as MAE3D), a novel autoencoding paradigm for self-supervised learning. We first split the input point cloud into patches and mask a portion of them, then use our Patch Embedding Module to extract the features of unmasked patches. Secondly, we employ patch-wise MAE3D Transformers to learn both local features of point cloud patches and high-level contextual relationships between patches and complete the latent representations of masked patches. We use our Point Cloud Reconstruction Module with multi-task loss to complete the incomplete point cloud as a result. We conduct self-supervised pre-training on ShapeNet55 with the point cloud completion pre-text task and fine-tune the pre-trained model on ModelNet40 and ScanObjectNN (PB\_T50\_RS, the hardest variant). Comprehensive experiments demonstrate that the local features extracted by our MAE3D from point cloud patches are beneficial for downstream classification tasks, soundly outperforming state-of-the-art methods ($93.4\%$ and $86.2\%$ classification accuracy, respectively)., Comment: Accepted to IEEE Transactions on Multimedia

Weakly Supervised Object Localization (WSOL) aims to localize objects with image-level supervision. Existing works mainly rely on Class Activation Mapping (CAM) derived from a classification model. However, CAM-based methods usually focus on the most discriminative parts of an object (i.e., incomplete localization problem). In this paper, we empirically prove that this problem is associated with the mixup of the activation values between less discriminative foreground regions and the background. To address it, we propose Class RE-Activation Mapping (CREAM), a novel clustering-based approach to boost the activation values of the integral object regions. To this end, we introduce class-specific foreground and background context embeddings as cluster centroids. A CAM-guided momentum preservation strategy is developed to learn the context embeddings during training. At the inference stage, the re-activation mapping is formulated as a parameter estimation problem under Gaussian Mixture Model, which can be solved by deriving an unsupervised Expectation-Maximization based soft-clustering algorithm. By simply integrating CREAM into various WSOL approaches, our method significantly improves their performance. CREAM achieves the state-of-the-art performance on CUB, ILSVRC and OpenImages benchmark datasets. Code will be available at https://github.com/Jazzcharles/CREAM., Comment: 10 pages, CVPR 2022

Histopathological image analysis is the gold standard to diagnose cancer. Carcinoma is a subtype of cancer that constitutes more than 80% of all cancer cases. Squamous cell carcinoma and adenocarcinoma are two major subtypes of carcinoma, diagnosed by microscopic study of biopsy slides. However, manual microscopic evaluation is a subjective and time-consuming process. Many researchers have reported methods to automate carcinoma detection and classification. The increasing use of artificial intelligence (AI) in the automation of carcinoma diagnosis also reveals a significant rise in the use of deep network models. In this systematic literature review, we present a comprehensive review of the state-of-the-art approaches reported in carcinoma diagnosis using histopathological images. Studies are selected from well-known databases with strict inclusion/exclusion criteria. We have categorized the articles and recapitulated their methods based on specific organs of carcinoma origin. Further, we have summarized pertinent literature on AI methods, highlighted critical challenges and limitations, and provided insights on future research direction in automated carcinoma diagnosis. Out of 101 articles selected, most of the studies experimented on private datasets with varied image sizes, obtaining accuracy between 63% and 100%. Overall, this review highlights the need for a generalized AI-based carcinoma diagnostic system. Additionally, it is desirable to have accountable approaches to extract microscopic features from images of multiple magnifications that should mimic pathologists' evaluations., Comment: accepted to Computers in Biology and Medicine

As a popular representation of 3D data, point cloud may contain noise and need to be filtered before use. Existing point cloud filtering methods either cannot preserve sharp features or result in uneven point distribution in the filtered output. To address this problem, this paper introduces a point cloud filtering method that considers both point distribution and feature preservation during filtering. The key idea is to incorporate a repulsion term with a data term in energy minimization. The repulsion term is responsible for the point distribution, while the data term is to approximate the noisy surfaces while preserving the geometric features. This method is capable of handling models with fine-scale features and sharp features. Extensive experiments show that our method yields better results with a more uniform point distribution ($5.8\times10^{-5}$ Chamfer Distance on average) in seconds., Comment: This paper is accepted to CVM

Intracranial aneurysms are common nowadays and how to detect them intelligently is of great significance in digital health. While most existing deep learning research focused on medical images in a supervised way, we introduce an unsupervised method for the detection of intracranial aneurysms based on 3D point cloud data. In particular, our method consists of two stages: unsupervised pre-training and downstream tasks. As for the former, the main idea is to pair each point cloud with its jittered counterpart and maximise their correspondence. Then we design a dual-branch contrastive network with an encoder for each branch and a subsequent common projection head. As for the latter, we design simple networks for supervised classification and segmentation training. Experiments on the public dataset (IntrA) show that our unsupervised method achieves comparable or even better performance than some state-of-the-art supervised techniques, and it is most prominent in the detection of aneurysmal vessels. Experiments on the ModelNet40 also show that our method achieves the accuracy of 90.79\% which outperforms existing state-of-the-art unsupervised models., Comment: under review (corresponding: {xuequan.lu@deakin.edu.au})

Normal estimation on 3D point clouds is a fundamental problem in 3D vision and graphics. Current methods often show limited accuracy in predicting normals at sharp features (e.g., edges and corners) and less robustness to noise. In this paper, we propose a novel normal estimation method for point clouds. It consists of two phases: (a) feature encoding which learns representations of local patches, and (b) normal estimation that takes the learned representation as input and regresses the normal vector. We are motivated that local patches on isotropic and anisotropic surfaces have similar or distinct normals, and that separable features or representations can be learned to facilitate normal estimation. To realise this, we first construct triplets of local patches on 3D point cloud data, and design a triplet network with a triplet loss for feature encoding. We then design a simple network with several MLPs and a loss function to regress the normal vector. Despite having a smaller network size compared to most other methods, experimental results show that our method preserves sharp features and achieves better normal estimation results on CAD-like shapes., Comment: Accepted by ICME 2022. Supplementary material available at https://ieeexplore.ieee.org/document/9859844/media#media

Existing position based point cloud filtering methods can hardly preserve sharp geometric features. In this paper, we rethink point cloud filtering from a non-learning non-local non-normal perspective, and propose a novel position based approach for feature-preserving point cloud filtering. Unlike normal based techniques, our method does not require the normal information. The core idea is to first design a similarity metric to search the non-local similar patches of a queried local patch. We then map the non-local similar patches into a canonical space and aggregate the non-local information. The aggregated outcome (i.e. coordinate) will be inversely mapped into the original space. Our method is simple yet effective. Extensive experiments validate our method, and show that it generally outperforms position based methods (deep learning and non-learning), and generates better or comparable outcomes to normal based techniques (deep learning and non-learning).

Though a number of point cloud learning methods have been proposed to handle unordered points, most of them are supervised and require labels for training. By contrast, unsupervised learning of point cloud data has received much less attention to date. In this paper, we propose a simple yet effective approach for unsupervised point cloud learning. In particular, we identify a very useful transformation which generates a good contrastive version of an original point cloud. They make up a pair. After going through a shared encoder and a shared head network, the consistency between the output representations are maximized with introducing two variants of contrastive losses to respectively facilitate downstream classification and segmentation. To demonstrate the efficacy of our method, we conduct experiments on three downstream tasks which are 3D object classification (on ModelNet40 and ModelNet10), shape part segmentation (on ShapeNet Part dataset) as well as scene segmentation (on S3DIS). Comprehensive results show that our unsupervised contrastive representation learning enables impressive outcomes in object classification and semantic segmentation. It generally outperforms current unsupervised methods, and even achieves comparable performance to supervised methods. Our source codes will be made publicly available.

Unsupervised domain adaptation (UDA) for semantic segmentation has been well-studied in recent years. However, most existing works largely neglect the local regional consistency across different domains and are less robust to changes in outdoor environments. In this paper, we propose a novel and fully end-to-end trainable approach, called regional contrastive consistency regularization (RCCR) for domain adaptive semantic segmentation. Our core idea is to pull the similar regional features extracted from the same location of different images, i.e., the original image and augmented image, to be closer, and meanwhile push the features from the different locations of the two images to be separated. We innovatively propose a region-wise contrastive loss with two sampling strategies to realize effective regional consistency. Besides, we present momentum projection heads, where the teacher projection head is the exponential moving average of the student. Finally, a memory bank mechanism is designed to learn more robust and stable region-wise features under varying environments. Extensive experiments on two common UDA benchmarks, i.e., GTAV to Cityscapes and SYNTHIA to Cityscapes, demonstrate that our approach outperforms the state-of-the-art methods., Comment: Accepted to IEEE International Conference on Multimedia and Expo (ICME), 2022

Most existing 3D geometry copy detection research focused on 3D watermarking, which first embeds ``watermarks'' and then detects the added watermarks. However, this kind of methods is non-straightforward and may be less robust to attacks such as cropping and noise. In this paper, we focus on a fundamental and practical research problem: judging whether a point cloud is plagiarized or copied to another point cloud in the presence of several manipulations (e.g., similarity transformation, smoothing). We propose a novel method to address this critical problem. Our key idea is first to align the two point clouds and then calculate their similarity distance. We design three different measures to compute the similarity. We also introduce two strategies to speed up our method. Comprehensive experiments and comparisons demonstrate the effectiveness and robustness of our method in estimating the similarity of two given 3D point clouds., Comment: submitted for review

3D anatomical landmarks play an important role in health research. Their automated prediction/localization thus becomes a vital task. In this paper, we introduce a deformation method for 3D anatomical landmarks prediction. It utilizes a source model with anatomical landmarks which are annotated by clinicians, and deforms this model non-rigidly to match the target model. Two constraints are introduced in the optimization, which are responsible for alignment and smoothness, respectively. Experiments are performed on our dataset and the results demonstrate the robustness of our method, and show that it yields better performance than the state-of-the-art techniques in most cases., Comment: submitted for review