Search

Neural Radiance Fields (NeRF) have advanced photorealistic novel view synthesis, but their reliance on photometric reconstruction introduces artifacts, commonly known as "floaters". These artifacts degrade novel view quality, especially in areas unseen by the training cameras. We present a fast, post-hoc NeRF cleanup method that eliminates such artifacts by enforcing our Free Space Prior, effectively minimizing floaters without disrupting the NeRF's representation of observed regions. Unlike existing approaches that rely on either Maximum Likelihood (ML) estimation to fit the data or a complex, local data-driven prior, our method adopts a Maximum-a-Posteriori (MAP) approach, selecting the optimal model parameters under a simple global prior assumption that unseen regions should remain empty. This enables our method to clean artifacts in both seen and unseen areas, enhancing novel view quality even in challenging scene regions. Our method is comparable with existing NeRF cleanup models while being 2.5x faster in inference time, requires no additional memory beyond the original NeRF, and achieves cleanup training in less than 30 seconds. Our code will be made publically available.

In the study of animal behavior, researchers often record long continuous videos, accumulating into large-scale datasets. However, the behaviors of interest are often rare compared to routine behaviors. This incurs a heavy cost on manual annotation, forcing users to sift through many samples before finding their needles. We propose a pipeline to efficiently sample rare behaviors from large datasets, enabling the creation of training datasets for rare behavior classifiers. Our method only needs an unlabeled animal pose or acceleration dataset as input and makes no assumptions regarding the type, number, or characteristics of the rare behaviors. Our pipeline is based on a recent graph-based anomaly detection model for human behavior, which we apply to this new data domain. It leverages anomaly scores to automatically label normal samples while directing human annotation efforts toward anomalies. In research data, anomalies may come from many different sources (e.g., signal noise versus true rare instances). Hence, the entire labeling budget is focused on the abnormal classes, letting the user review and label samples according to their needs. We tested our approach on three datasets of freely-moving animals, acquired in the laboratory and the field. We found that graph-based models are particularly useful when studying motion-based behaviors in animals, yielding good results while using a small labeling budget. Our method consistently outperformed traditional random sampling, offering an average improvement of 70% in performance and creating datasets even when the behavior of interest was only 0.02% of the data. Even when the performance gain was minor (e.g., when the behavior is not rare), our method still reduced the annotation effort by half.

Diffusion models dominate the space of text-to-image generation, yet they may produce undesirable outputs, including explicit content or private data. To mitigate this, concept ablation techniques have been explored to limit the generation of certain concepts. In this paper, we reveal that the erased concept information persists in the model and that erased concept images can be generated using the right latent. Utilizing inversion methods, we show that there exist latent seeds capable of generating high quality images of erased concepts. Moreover, we show that these latents have likelihoods that overlap with those of images outside the erased concept. We extend this to demonstrate that for every image from the erased concept set, we can generate many seeds that generate the erased concept. Given the vast space of latents capable of generating ablated concept images, our results suggest that fully erasing concept information may be intractable, highlighting possible vulnerabilities in current concept ablation techniques., Comment: The first three authors contributed equally to this work. Project page: https://matanr.github.io/Memories_of_Forgotten_Concepts/

The longstanding Godbersen's conjecture states that for any convex body $K \subset \mathbb R^n$ of volume $1$ and any $j \in \{0, \ldots, n\}$, the mixed volume $V_j = V(K[j], -K[n - j])$ is bounded by $\binom{n}{j}$, with equality if and only if $K$ is a simplex. We demonstrate that several consequences of this conjecture are true: certain families of linear combinations of the $V_j$, arising from different geometric constructions, are bounded above by their values when one substitutes $\binom{n}{j}$ for $V_j$, with equality if and only if $K$ is a simplex. One of our results implies that for any $K$ of volume $1$ we have $\frac{1}{n + 1} \sum_{j = 0}^n \binom{n}{j}^{-1} V_j \le 1$, showing that Godbersen's conjecture holds ``on average'' for any body. Another result generalizes the well-known Rogers-Shephard inequality for the difference body., Comment: 13 pages. arXiv admin note: text overlap with arXiv:1703.06403

Category-Agnostic Pose Estimation (CAPE) localizes keypoints across diverse object categories with a single model, using one or a few annotated support images. Recent works have shown that using a pose graph (i.e., treating keypoints as nodes in a graph rather than isolated points) helps handle occlusions and break symmetry. However, these methods assume a static pose graph with equal-weight edges, leading to suboptimal results. We introduce EdgeCape, a novel framework that overcomes these limitations by predicting the graph's edge weights which optimizes localization. To further leverage structural priors, we propose integrating Markovian Structural Bias, which modulates the self-attention interaction between nodes based on the number of hops between them. We show that this improves the model's ability to capture global spatial dependencies. Evaluated on the MP-100 benchmark, which includes 100 categories and over 20K images, EdgeCape achieves state-of-the-art results in the 1-shot setting and leads among similar-sized methods in the 5-shot setting, significantly improving keypoint localization accuracy. Our code is publicly available.

In this paper we define and investigate a class of polytopes which we call "vertex generated" consisting of polytopes which are the average of their $0$ and $n$ dimensional faces. We show many results regarding this class, among them: that the class contains all zonotopes, that it is dense in dimension $n=2$, that any polytope can be summed with a zonotope so that the sum is in this class, and that a strong form of the celebrated "Maurey Lemma" holds for polytopes in this class. We introduce for every polytope a parameter which measures how far it is from being vertex-generated, and show that when this parameter is small, strong covering properties hold., Comment: replacement of 2306.15293, paper no. 2/3. 22 pages, 2 figures

Conventional 2D pose estimation models are constrained by their design to specific object categories. This limits their applicability to predefined objects. To overcome these limitations, category-agnostic pose estimation (CAPE) emerged as a solution. CAPE aims to facilitate keypoint localization for diverse object categories using a unified model, which can generalize from minimal annotated support images. Recent CAPE works have produced object poses based on arbitrary keypoint definitions annotated on a user-provided support image. Our work departs from conventional CAPE methods, which require a support image, by adopting a text-based approach instead of the support image. Specifically, we use a pose-graph, where nodes represent keypoints that are described with text. This representation takes advantage of the abstraction of text descriptions and the structure imposed by the graph. Our approach effectively breaks symmetry, preserves structure, and improves occlusion handling. We validate our novel approach using the MP-100 benchmark, a comprehensive dataset spanning over 100 categories and 18,000 images. Under a 1-shot setting, our solution achieves a notable performance boost of 1.07\%, establishing a new state-of-the-art for CAPE. Additionally, we enrich the dataset by providing text description annotations, further enhancing its utility for future research.

Curriculum learning has been growing in the domain of reinforcement learning as a method of improving training efficiency for various tasks. It involves modifying the difficulty (lessons) of the environment as the agent learns, in order to encourage more optimal agent behavior and higher reward states. However, most curriculum learning methods currently involve discrete transitions of the curriculum or predefined steps by the programmer or using automatic curriculum learning on only a small subset training such as only on an adversary. In this paper, we propose a novel approach to curriculum learning that uses a Setter Model to automatically generate an action space, adversary strength, initialization, and bunching strength. Transportation and traffic optimization is a well known area of study, especially for reinforcement learning based solutions. We specifically look at the bus bunching problem for the context of this study. The main idea of the problem is to minimize the delays caused by inefficient bus timings for passengers arriving and departing from a system of buses. While the heavy exploration in the area makes innovation and improvement with regards to performance marginal, it simultaneously provides an effective baseline for developing new generalized techniques. Our group is particularly interested in examining curriculum learning and its effect on training efficiency and overall performance. We decide to try a lesser known approach to curriculum learning, in which the curriculum is not fixed or discretely thresholded. Our method for automated curriculum learning involves a curriculum that is dynamically chosen and learned by an adversary network made to increase the difficulty of the agent's training, and defined by multiple forms of input. Our results are shown in the following sections of this paper., Comment: 9 pages, preprint

3D scene registration is a fundamental problem in computer vision that seeks the best 6-DoF alignment between two scenes. This problem was extensively investigated in the case of point clouds and meshes, but there has been relatively limited work regarding Neural Radiance Fields (NeRF). In this paper, we consider the problem of rigid registration between two NeRFs when the position of the original cameras is not given. Our key novelty is the introduction of Viewshed Fields (VF), an implicit function that determines, for each 3D point, how likely it is to be viewed by the original cameras. We demonstrate how VF can help in the various stages of NeRF registration, with an extensive evaluation showing that VF-NeRF achieves SOTA results on various datasets with different capturing approaches such as LLFF and Objaverese.

Diffusion inversion is the problem of taking an image and a text prompt that describes it and finding a noise latent that would generate the exact same image. Most current deterministic inversion techniques operate by approximately solving an implicit equation and may converge slowly or yield poor reconstructed images. We formulate the problem by finding the roots of an implicit equation and devlop a method to solve it efficiently. Our solution is based on Newton-Raphson (NR), a well-known technique in numerical analysis. We show that a vanilla application of NR is computationally infeasible while naively transforming it to a computationally tractable alternative tends to converge to out-of-distribution solutions, resulting in poor reconstruction and editing. We therefore derive an efficient guided formulation that fastly converges and provides high-quality reconstructions and editing. We showcase our method on real image editing with three popular open-sourced diffusion models: Stable Diffusion, SDXL-Turbo, and Flux with different deterministic schedulers. Our solution, Guided Newton-Raphson Inversion, inverts an image within 0.4 sec (on an A100 GPU) for few-step models (SDXL-Turbo and Flux.1), opening the door for interactive image editing. We further show improved results in image interpolation and generation of rare objects.

Vector image representation is a popular choice when editability and flexibility in resolution are desired. However, most images are only available in raster form, making raster-to-vector image conversion (vectorization) an important task. Classical methods for vectorization are either domain-specific or yield an abundance of shapes which limits editability and interpretability. Learning-based methods, that use differentiable rendering, have revolutionized vectorization, at the cost of poor generalization to out-of-training distribution domains, and optimization-based counterparts are either slow or produce non-editable and redundant shapes. In this work, we propose Optimize & Reduce (O&R), a top-down approach to vectorization that is both fast and domain-agnostic. O&R aims to attain a compact representation of input images by iteratively optimizing B\'ezier curve parameters and significantly reducing the number of shapes, using a devised importance measure. We contribute a benchmark of five datasets comprising images from a broad spectrum of image complexities - from emojis to natural-like images. Through extensive experiments on hundreds of images, we demonstrate that our method is domain agnostic and outperforms existing works in both reconstruction and perceptual quality for a fixed number of shapes. Moreover, we show that our algorithm is $\times 10$ faster than the state-of-the-art optimization-based method.

In this paper we explore questions regarding the Minkowski sum of the boundaries of convex sets. Motivated by a question suggested to us by V.~Milman regarding the volume of $\partial K+ \partial T$ where $K$ and $T$ are convex bodies, we prove sharp volumetric lower bounds for the Minkowski average of the boundaries of sets with connected boundary, as well as some related results., Comment: Previous version extended and split into three papers. This is paper no. 1; 13 pages, 3 figures, published in Communications in Contemporary Mathematics

MLaaS Service Providers (SPs) holding a Neural Network would like to keep the Neural Network weights secret. On the other hand, users wish to utilize the SPs' Neural Network for inference without revealing their data. Multi-Party Computation (MPC) offers a solution to achieve this. Computations in MPC involve communication, as the parties send data back and forth. Non-linear operations are usually the main bottleneck requiring the bulk of communication bandwidth. In this paper, we focus on ResNets, which serve as the backbone for many Computer Vision tasks, and we aim to reduce their non-linear components, specifically, the number of ReLUs. Our key insight is that spatially close pixels exhibit correlated ReLU responses. Building on this insight, we replace the per-pixel ReLU operation with a ReLU operation per patch. We term this approach 'Block-ReLU'. Since different layers in a Neural Network correspond to different feature hierarchies, it makes sense to allow patch-size flexibility for the various layers of the Neural Network. We devise an algorithm to choose the optimal set of patch sizes through a novel reduction of the problem to the Knapsack Problem. We demonstrate our approach in the semi-honest secure 3-party setting for four problems: Classifying ImageNet using ResNet50 backbone, classifying CIFAR100 using ResNet18 backbone, Semantic Segmentation of ADE20K using MobileNetV2 backbone, and Semantic Segmentation of Pascal VOC 2012 using ResNet50 backbone. Our approach achieves competitive performance compared to a handful of competitors. Our source code is publicly available: https://github.com/yg320/secure_inference.

We propose SampleDepth, a Convolutional Neural Network (CNN), that is suited for an adaptive LiDAR. Typically,LiDAR sampling strategy is pre-defined, constant and independent of the observed scene. Instead of letting a LiDAR sample the scene in this agnostic fashion, SampleDepth determines, adaptively, where it is best to sample the current frame. To do that, SampleDepth uses depth samples from previous time steps to predict a sampling mask for the current frame. Crucially, SampleDepth is trained to optimize the performance of a depth completion downstream task. SampleDepth is evaluated on two different depth completion networks and two LiDAR datasets, KITTI Depth Completion and the newly introduced synthetic dataset, SHIFT. We show that SampleDepth is effective and suitable for different depth completion downstream tasks.

We propose an algorithm for taming Normalizing Flow models - changing the probability that the model will produce a specific image or image category. We focus on Normalizing Flows because they can calculate the exact generation probability likelihood for a given image. We demonstrate taming using models that generate human faces, a subdomain with many interesting privacy and bias considerations. Our method can be used in the context of privacy, e.g., removing a specific person from the output of a model, and also in the context of debiasing by forcing a model to output specific image categories according to a given target distribution. Taming is achieved with a fast fine-tuning process without retraining the model from scratch, achieving the goal in a matter of minutes. We evaluate our method qualitatively and quantitatively, showing that the generation quality remains intact, while the desired changes are applied.

Scene flow estimation is a long-standing problem in computer vision, where the goal is to find the 3D motion of a scene from its consecutive observations. Recently, there have been efforts to compute the scene flow from 3D point clouds. A common approach is to train a regression model that consumes source and target point clouds and outputs the per-point translation vector. An alternative is to learn point matches between the point clouds concurrently with regressing a refinement of the initial correspondence flow. In both cases, the learning task is very challenging since the flow regression is done in the free 3D space, and a typical solution is to resort to a large annotated synthetic dataset. We introduce SCOOP, a new method for scene flow estimation that can be learned on a small amount of data without employing ground-truth flow supervision. In contrast to previous work, we train a pure correspondence model focused on learning point feature representation and initialize the flow as the difference between a source point and its softly corresponding target point. Then, in the run-time phase, we directly optimize a flow refinement component with a self-supervised objective, which leads to a coherent and accurate flow field between the point clouds. Experiments on widespread datasets demonstrate the performance gains achieved by our method compared to existing leading techniques while using a fraction of the training data. Our code is publicly available at https://github.com/itailang/SCOOP., Comment: CVPR 2023. Project page: https://itailang.github.io/SCOOP/

A triangular mesh is one of the most popular 3D data representations. As such, the deployment of deep neural networks for mesh processing is widely spread and is increasingly attracting more attention. However, neural networks are prone to adversarial attacks, where carefully crafted inputs impair the model's functionality. The need to explore these vulnerabilities is a fundamental factor in the future development of 3D-based applications. Recently, mesh attacks were studied on the semantic level, where classifiers are misled to produce wrong predictions. Nevertheless, mesh surfaces possess complex geometric attributes beyond their semantic meaning, and their analysis often includes the need to encode and reconstruct the geometry of the shape. We propose a novel framework for a geometric adversarial attack on a 3D mesh autoencoder. In this setting, an adversarial input mesh deceives the autoencoder by forcing it to reconstruct a different geometric shape at its output. The malicious input is produced by perturbing a clean shape in the spectral domain. Our method leverages the spectral decomposition of the mesh along with additional mesh-related properties to obtain visually credible results that consider the delicacy of surface distortions. Our code is publicly available at https://github.com/StolikTomer/SAGA., Comment: Presented at ICCV 2023. Project page: https://stoliktomer.github.io/SAGA/

Video anomaly detection is an ill-posed problem because it relies on many parameters such as appearance, pose, camera angle, background, and more. We distill the problem to anomaly detection of human pose, thus decreasing the risk of nuisance parameters such as appearance affecting the result. Focusing on pose alone also has the side benefit of reducing bias against distinct minority groups. Our model works directly on human pose graph sequences and is exceptionally lightweight (~1K parameters), capable of running on any machine able to run the pose estimation with negligible additional resources. We leverage the highly compact pose representation in a normalizing flows framework, which we extend to tackle the unique characteristics of spatio-temporal pose data and show its advantages in this use case. The algorithm is quite general and can handle training data of only normal examples as well as a supervised setting that consists of labeled normal and abnormal examples. We report state-of-the-art results on two anomaly detection benchmarks - the unsupervised ShanghaiTech dataset and the recent supervised UBnormal dataset.

Although prediction models for delirium, a commonly occurring condition during general hospitalization or post-surgery, have not gained huge popularity, their algorithmic bias evaluation is crucial due to the existing association between social determinants of health and delirium risk. In this context, using MIMIC-III and another academic hospital dataset, we present some initial experimental evidence showing how sociodemographic features such as sex and race can impact the model performance across subgroups. With this work, our intent is to initiate a discussion about the intersectionality effects of old age, race and socioeconomic factors on the early-stage detection and prevention of delirium using ML., Comment: Extended Abstract presented at Machine Learning for Health (ML4H) symposium 2022, November 28th, 2022, New Orleans, United States & Virtual, http://www.ml4h.cc, 14 pages

The increasing popularity of facial manipulation (Deepfakes) and synthetic face creation raises the need to develop robust forgery detection solutions. Crucially, most work in this domain assume that the Deepfakes in the test set come from the same Deepfake algorithms that were used for training the network. This is not how things work in practice. Instead, we consider the case where the network is trained on one Deepfake algorithm, and tested on Deepfakes generated by another algorithm. Typically, supervised techniques follow a pipeline of visual feature extraction from a deep backbone, followed by a binary classification head. Instead, our algorithm aggregates features extracted across all layers of one backbone network to detect a fake. We evaluate our approach on two domains of interest - Deepfake detection and Synthetic image detection, and find that we achieve SOTA results.

Major postoperative complications are devastating to surgical patients. Some of these complications are potentially preventable via early predictions based on intraoperative data. However, intraoperative data comprise long and fine-grained multivariate time series, prohibiting the effective learning of accurate models. The large gaps associated with clinical events and protocols are usually ignored. Moreover, deep models generally lack transparency. Nevertheless, the interpretability is crucial to assist clinicians in planning for and delivering postoperative care and timely interventions. Towards this end, we propose a hierarchical model combining the strength of both attention and recurrent models for intraoperative time series. We further develop an explanation module for the hierarchical model to interpret the predictions by providing contributions of intraoperative data in a fine-grained manner. Experiments on a large dataset of 111,888 surgeries with multiple outcomes and an external high-resolution ICU dataset show that our model can achieve strong predictive performance (i.e., high accuracy) and offer robust interpretations (i.e., high transparency) for predicted outcomes based on intraoperative time series.

An applied problem facing all areas of data science is harmonizing data sources. Joining data from multiple origins with unmapped and only partially overlapping features is a prerequisite to developing and testing robust, generalizable algorithms, especially in health care. We approach this issue in the common but difficult case of numeric features such as nearly Gaussian and binary features, where unit changes and variable shift make simple matching of univariate summaries unsuccessful. We develop two novel procedures to address this problem. First, we demonstrate multiple methods of "fingerprinting" a feature based on its associations to other features. In the setting of even modest prior information, this allows most shared features to be accurately identified. Second, we demonstrate a deep learning algorithm for translation between databases. Unlike prior approaches, our algorithm takes advantage of discovered mappings while identifying surrogates for unshared features and learning transformations. In synthetic and real-world experiments using two electronic health record databases, our algorithms outperform existing baselines for matching variable sets, while jointly learning to impute unshared or transformed variables.

Rigid Point Cloud Registration (PCR) algorithms aim to estimate the 6-DOF relative motion between two point clouds, which is important in various fields, including autonomous driving. Recent years have seen a significant improvement in global PCR algorithms, i.e. algorithms that can handle a large relative motion. This has been demonstrated in various scenarios, including indoor scenes, but has only been minimally tested in the Automotive setting, where point clouds are produced by vehicle-mounted LiDAR sensors. In this work, we aim to answer questions that are important for automotive applications, including: which of the new algorithms is the most accurate, and which is fastest? How transferable are deep-learning approaches, e.g. what happens when you train a network with data from Boston, and run it in a vehicle in Singapore? How small can the overlap between point clouds be before the algorithms start to deteriorate? To what extent are the algorithms rotation invariant? Our results are at times surprising. When comparing robust parameter estimation methods for registration, we find that the fastest and most accurate is not one of the newest approaches. Instead, it is a modern variant of the well known RANSAC technique. We also suggest a new outlier filtering method, Grid-Prioritized Filtering (GPF), to further improve it. An additional contribution of this work is an algorithm for selecting challenging sets of frame-pairs from automotive LiDAR datasets. This enables meaningful benchmarking in the Automotive LiDAR setting, and can also improve training for learning algorithms., Comment: Accepted to the NeurIPS 2022 workshop on Machine Learning for Autonomous Driving. Project Page: https://github.com/AmnonDrory/LidarRegistration