Search

Generative models have become a powerful tool for image editing tasks, including object insertion. However, these methods often lack spatial awareness, generating objects with unrealistic locations and scales, or unintentionally altering the scene background. A key challenge lies in maintaining visual coherence, which requires both a geometrically suitable object location and a high-quality image edit. In this paper, we focus on the former, creating a location model dedicated to identifying realistic object locations. Specifically, we train an autoregressive model that generates bounding box coordinates, conditioned on the background image and the desired object class. This formulation allows to effectively handle sparse placement annotations and to incorporate implausible locations into a preference dataset by performing direct preference optimization. Our extensive experiments demonstrate that our generative location model, when paired with an inpainting method, substantially outperforms state-of-the-art instruction-tuned models and location modeling baselines in object insertion tasks, delivering accurate and visually coherent results.

Robot arms should be able to learn new tasks. One framework here is reinforcement learning, where the robot is given a reward function that encodes the task, and the robot autonomously learns actions to maximize its reward. Existing approaches to reinforcement learning often frame this problem as a Markov decision process, and learn a policy (or a hierarchy of policies) to complete the task. These policies reason over hundreds of fine-grained actions that the robot arm needs to take: e.g., moving slightly to the right or rotating the end-effector a few degrees. But the manipulation tasks that we want robots to perform can often be broken down into a small number of high-level motions: e.g., reaching an object or turning a handle. In this paper we therefore propose a waypoint-based approach for model-free reinforcement learning. Instead of learning a low-level policy, the robot now learns a trajectory of waypoints, and then interpolates between those waypoints using existing controllers. Our key novelty is framing this waypoint-based setting as a sequence of multi-armed bandits: each bandit problem corresponds to one waypoint along the robot's motion. We theoretically show that an ideal solution to this reformulation has lower regret bounds than standard frameworks. We also introduce an approximate posterior sampling solution that builds the robot's motion one waypoint at a time. Results across benchmark simulations and two real-world experiments suggest that this proposed approach learns new tasks more quickly than state-of-the-art baselines. See videos here: https://youtu.be/MMEd-lYfq4Y

Diffusion-based video editing have reached impressive quality and can transform either the global style, local structure, and attributes of given video inputs, following textual edit prompts. However, such solutions typically incur heavy memory and computational costs to generate temporally-coherent frames, either in the form of diffusion inversion and/or cross-frame attention. In this paper, we conduct an analysis of such inefficiencies, and suggest simple yet effective modifications that allow significant speed-ups whilst maintaining quality. Moreover, we introduce Object-Centric Diffusion, to fix generation artifacts and further reduce latency by allocating more computations towards foreground edited regions, arguably more important for perceptual quality. We achieve this by two novel proposals: i) Object-Centric Sampling, decoupling the diffusion steps spent on salient or background regions and spending most on the former, and ii) Object-Centric Token Merging, which reduces cost of cross-frame attention by fusing redundant tokens in unimportant background regions. Both techniques are readily applicable to a given video editing model without retraining, and can drastically reduce its memory and computational cost. We evaluate our proposals on inversion-based and control-signal-based editing pipelines, and show a latency reduction up to 10x for a comparable synthesis quality. Project page: qualcomm-ai-research.github.io/object-centric-diffusion., Comment: ECCV24

Novel View Synthesis (NVS), which tries to produce a realistic image at the target view given source view images and their corresponding poses, is a fundamental problem in 3D Vision. As this task is heavily under-constrained, some recent work, like Zero123, tries to solve this problem with generative modeling, specifically using pre-trained diffusion models. Although this strategy generalizes well to new scenes, compared to neural radiance field-based methods, it offers low levels of flexibility. For example, it can only accept a single-view image as input, despite realistic applications often offering multiple input images. This is because the source-view images and corresponding poses are processed separately and injected into the model at different stages. Thus it is not trivial to generalize the model into multi-view source images, once they are available. To solve this issue, we try to process each pose image pair separately and then fuse them as a unified visual representation which will be injected into the model to guide image synthesis at the target-views. However, inconsistency and computation costs increase as the number of input source-view images increases. To solve these issues, the Multi-view Cross Former module is proposed which maps variable-length input data to fix-size output data. A two-stage training strategy is introduced to further improve the efficiency during training time. Qualitative and quantitative evaluation over multiple datasets demonstrates the effectiveness of the proposed method against previous approaches. The code will be released according to the acceptance., Comment: paper and supplementary material

This work aims to improve the efficiency of text-to-image diffusion models. While diffusion models use computationally expensive UNet-based denoising operations in every generation step, we identify that not all operations are equally relevant for the final output quality. In particular, we observe that UNet layers operating on high-res feature maps are relatively sensitive to small perturbations. In contrast, low-res feature maps influence the semantic layout of the final image and can often be perturbed with no noticeable change in the output. Based on this observation, we propose Clockwork Diffusion, a method that periodically reuses computation from preceding denoising steps to approximate low-res feature maps at one or more subsequent steps. For multiple baselines, and for both text-to-image generation and image editing, we demonstrate that Clockwork leads to comparable or improved perceptual scores with drastically reduced computational complexity. As an example, for Stable Diffusion v1.5 with 8 DPM++ steps we save 32% of FLOPs with negligible FID and CLIP change.

For robots to seamlessly interact with humans, we first need to make sure that humans and robots understand one another. Diverse algorithms have been developed to enable robots to learn from humans (i.e., transferring information from humans to robots). In parallel, visual, haptic, and auditory communication interfaces have been designed to convey the robot's internal state to the human (i.e., transferring information from robots to humans). Prior research often separates these two directions of information transfer, and focuses primarily on either learning algorithms or communication interfaces. By contrast, in this review we take an interdisciplinary approach to identify common themes and emerging trends that close the loop between learning and communication. Specifically, we survey state-of-the-art methods and outcomes for communicating a robot's learning back to the human teacher during human-robot interaction. This discussion connects human-in-the-loop learning methods and explainable robot learning with multi-modal feedback systems and measures of human-robot interaction. We find that -- when learning and communication are developed together -- the resulting closed-loop system can lead to improved human teaching, increased human trust, and human-robot co-adaptation. The paper includes a perspective on several of the interdisciplinary research themes and open questions that could advance how future robots communicate their learning to everyday operators. Finally, we implement a selection of the reviewed methods in a case study where participants kinesthetically teach a robot arm. This case study documents and tests an integrated approach for learning in ways that can be communicated, conveying this learning across multi-modal interfaces, and measuring the resulting changes in human and robot behavior. See videos of our case study here: https://youtu.be/EXfQctqFzWs

This paper accelerates video perception, such as semantic segmentation and human pose estimation, by levering cross-frame redundancies. Unlike the existing approaches, which avoid redundant computations by warping the past features using optical-flow or by performing sparse convolutions on frame differences, we approach the problem from a new perspective: low-bit quantization. We observe that residuals, as the difference in network activations between two neighboring frames, exhibit properties that make them highly quantizable. Based on this observation, we propose a novel quantization scheme for video networks coined as Residual Quantization. ResQ extends the standard, frame-by-frame, quantization scheme by incorporating temporal dependencies that lead to better performance in terms of accuracy vs. bit-width. Furthermore, we extend our model to dynamically adjust the bit-width proportional to the amount of changes in the video. We demonstrate the superiority of our model, against the standard quantization and existing efficient video perception models, using various architectures on semantic segmentation and human pose estimation benchmarks., Comment: ICCV 2023

Objective: Lung ultrasonography is a significant advance toward a harmless lung imagery system. This work has investigated the automatic localization of diagnostically significant features in lung ultrasound pictures which are Pleural line, A-lines, and B-lines. Study Design: Wavelet and Radon transforms have been utilized in order to denoise and highlight the presence of clinically significant patterns. The proposed framework is developed and validated using three different lung ultrasound image datasets. Two of them contain synthetic data and the other one is taken from the publicly available POCUS dataset. The efficiency of the proposed method is evaluated using 200 real images. Results: The obtained results prove that the comparison between localized patterns and the baselines yields a promising F2-score of 62%, 86%, and 100% for B-lines, A-lines, and Pleural line, respectively. Conclusion: Finally, the high F-scores attained show that the developed technique is an effective way to automatically extract lung patterns from ultrasound images., Comment: 13 pages, 12 figures, 4 tables

This work aims to improve the efficiency of vision transformers (ViT). While ViTs use computationally expensive self-attention operations in every layer, we identify that these operations are highly correlated across layers -- a key redundancy that causes unnecessary computations. Based on this observation, we propose SkipAt, a method to reuse self-attention computation from preceding layers to approximate attention at one or more subsequent layers. To ensure that reusing self-attention blocks across layers does not degrade the performance, we introduce a simple parametric function, which outperforms the baseline transformer's performance while running computationally faster. We show the effectiveness of our method in image classification and self-supervised learning on ImageNet-1K, semantic segmentation on ADE20K, image denoising on SIDD, and video denoising on DAVIS. We achieve improved throughput at the same-or-higher accuracy levels in all these tasks.

Humans can leverage physical interaction to teach robot arms. As the human kinesthetically guides the robot through demonstrations, the robot learns the desired task. While prior works focus on how the robot learns, it is equally important for the human teacher to understand what their robot is learning. Visual displays can communicate this information; however, we hypothesize that visual feedback alone misses out on the physical connection between the human and robot. In this paper we introduce a novel class of soft haptic displays that wrap around the robot arm, adding signals without affecting that interaction. We first design a pneumatic actuation array that remains flexible in mounting. We then develop single and multi-dimensional versions of this wrapped haptic display, and explore human perception of the rendered signals during psychophysic tests and robot learning. We ultimately find that people accurately distinguish single-dimensional feedback with a Weber fraction of 11.4%, and identify multi-dimensional feedback with 94.5% accuracy. When physically teaching robot arms, humans leverage the single- and multi-dimensional feedback to provide better demonstrations than with visual feedback: our wrapped haptic display decreases teaching time while increasing demonstration quality. This improvement depends on the location and distribution of the wrapped haptic display. You can see videos of our device and experiments here: https://youtu.be/yPcMGeqsjdM, Comment: This article includes work originally presented in arXiv:2111.04542

Though the state-of-the architectures for semantic segmentation, such as HRNet, demonstrate impressive accuracy, the complexity arising from their salient design choices hinders a range of model acceleration tools, and further they make use of operations that are inefficient on current hardware. This paper demonstrates that a simple encoder-decoder architecture with a ResNet-like backbone and a small multi-scale head, performs on-par or better than complex semantic segmentation architectures such as HRNet, FANet and DDRNets. Naively applying deep backbones designed for Image Classification to the task of Semantic Segmentation leads to sub-par results, owing to a much smaller effective receptive field of these backbones. Implicit among the various design choices put forth in works like HRNet, DDRNet, and FANet are networks with a large effective receptive field. It is natural to ask if a simple encoder-decoder architecture would compare favorably if comprised of backbones that have a larger effective receptive field, though without the use of inefficient operations like dilated convolutions. We show that with minor and inexpensive modifications to ResNets, enlarging the receptive field, very simple and competitive baselines can be created for Semantic Segmentation. We present a family of such simple architectures for desktop as well as mobile targets, which match or exceed the performance of complex models on the Cityscapes dataset. We hope that our work provides simple yet effective baselines for practitioners to develop efficient semantic segmentation models., Comment: To be presented at Efficient Deep Learning for Computer Vision Workshop at CVPR 2022

Recent works explore collaboration between humans and teams of robots. These approaches make sense if the human is already working with the robot team; but how should robots encourage nearby humans to join their teams in the first place? Inspired by behavioral economics, we recognize that humans care about more than just team efficiency -- humans also have biases and expectations for team dynamics. Our hypothesis is that the way inclusive robots divide the task (i.e., how the robots split a larger task into subtask allocations) should be both legible and fair to the human partner. In this paper we introduce a bilevel optimization approach that enables robot teams to identify high-level subtask allocations and low-level trajectories that optimize for legibility, fairness, or a combination of both objectives. We then test our resulting algorithm across studies where humans watch or play with robot teams. We find that our approach to generating legible teams makes the human's role clear, and that humans typically prefer to join and collaborate with legible teams instead of teams that only optimize for efficiency. Incorporating fairness alongside legibility further encourages participation: when humans play with robots, we find that they prefer (potentially inefficient) teams where the subtasks or effort are evenly divided. See videos of our studies here https://youtu.be/cfN7O5na3mg

High-resolution images are widely adopted for high-performance object detection in videos. However, processing high-resolution inputs comes with high computation costs, and naive down-sampling of the input to reduce the computation costs quickly degrades the detection performance. In this paper, we propose SALISA, a novel non-uniform SALiency-based Input SAmpling technique for video object detection that allows for heavy down-sampling of unimportant background regions while preserving the fine-grained details of a high-resolution image. The resulting image is spatially smaller, leading to reduced computational costs while enabling a performance comparable to a high-resolution input. To achieve this, we propose a differentiable resampling module based on a thin plate spline spatial transformer network (TPS-STN). This module is regularized by a novel loss to provide an explicit supervision signal to learn to "magnify" salient regions. We report state-of-the-art results in the low compute regime on the ImageNet-VID and UA-DETRAC video object detection datasets. We demonstrate that on both datasets, the mAP of an EfficientDet-D1 (EfficientDet-D2) gets on par with EfficientDet-D2 (EfficientDet-D3) at a much lower computational cost. We also show that SALISA significantly improves the detection of small objects. In particular, SALISA with an EfficientDet-D1 detector improves the detection of small objects by $77\%$, and remarkably also outperforms EfficientDetD3 baseline., Comment: 20 pages, 7 figures

When robots interact with human partners, often these partners change their behavior in response to the robot. On the one hand this is challenging because the robot must learn to coordinate with a dynamic partner. But on the other hand -- if the robot understands these dynamics -- it can harness its own behavior, influence the human, and guide the team towards effective collaboration. Prior research enables robots to learn to influence other robots or simulated agents. In this paper we extend these learning approaches to now influence humans. What makes humans especially hard to influence is that -- not only do humans react to the robot -- but the way a single user reacts to the robot may change over time, and different humans will respond to the same robot behavior in different ways. We therefore propose a robust approach that learns to influence changing partner dynamics. Our method first trains with a set of partners across repeated interactions, and learns to predict the current partner's behavior based on the previous states, actions, and rewards. Next, we rapidly adapt to new partners by sampling trajectories the robot learned with the original partners, and then leveraging those existing behaviors to influence the new partner dynamics. We compare our resulting algorithm to state-of-the-art baselines across simulated environments and a user study where the robot and participants collaborate to build towers. We find that our approach outperforms the alternatives, even when the partner follows new or unexpected dynamics. Videos of the user study are available here: https://youtu.be/lYsWM8An18g

This paper aims to accelerate video stream processing, such as object detection and semantic segmentation, by leveraging the temporal redundancies that exist between video frames. Instead of propagating and warping features using motion alignment, such as optical flow, we propose a novel knowledge distillation schema coined as Delta Distillation. In our proposal, the student learns the variations in the teacher's intermediate features over time. We demonstrate that these temporal variations can be effectively distilled due to the temporal redundancies within video frames. During inference, both teacher and student cooperate for providing predictions: the former by providing initial representations extracted only on the key-frame, and the latter by iteratively estimating and applying deltas for the successive frames. Moreover, we consider various design choices to learn optimal student architectures including an end-to-end learnable architecture search. By extensive experiments on a wide range of architectures, including the most efficient ones, we demonstrate that delta distillation sets a new state of the art in terms of accuracy vs. efficiency trade-off for semantic segmentation and object detection in videos. Finally, we show that, as a by-product, delta distillation improves the temporal consistency of the teacher model.

Humans do not perceive all parts of a scene with the same resolution, but rather focus on few regions of interest (ROIs). Traditional Object-Based codecs take advantage of this biological intuition, and are capable of non-uniform allocation of bits in favor of salient regions, at the expense of increased distortion the remaining areas: such a strategy allows a boost in perceptual quality under low rate constraints. Recently, several neural codecs have been introduced for video compression, yet they operate uniformly over all spatial locations, lacking the capability of ROI-based processing. In this paper, we introduce two models for ROI-based neural video coding. First, we propose an implicit model that is fed with a binary ROI mask and it is trained by de-emphasizing the distortion of the background. Secondly, we design an explicit latent scaling method, that allows control over the quantization binwidth for different spatial regions of latent variables, conditioned on the ROI mask. By extensive experiments, we show that our methods outperform all our baselines in terms of Rate-Distortion (R-D) performance in the ROI. Moreover, they can generalize to different datasets and to any arbitrary ROI at inference time. Finally, they do not require expensive pixel-level annotations during training, as synthetic ROI masks can be used with little to no degradation in performance. To the best of our knowledge, our proposals are the first solutions that integrate ROI-based capabilities into neural video compression models., Comment: Updated arxiv version to the camera-ready version after acceptance at British Machine Vision Conference (BMVC) 2022

This study aims to examine the impact of electronic word-of-mouth (eWoM) marketing on branding attitudes in social networks. Specifically, we investigate the effects of eWoM activities on brand awareness, brand destruction, branding, brand image, and brand competition. To gather data, we conducted a survey among followers of the Vizland shoe page on the Instagram social network. Statistical analysis was performed using SPSS software, and hypotheses were tested using SmartPLS software. The results indicate that eWoM significantly and positively influences branding, brand image, and brand awareness, while it does not have an impact on brand destruction. Additionally, branding and brand destruction are found to play a crucial role in gaining competitive advantage. Therefore, eWoM activities contribute to enhancing the brand relationship.

Roboticists have been seeking to address this situation in recent years through the use of soft robots. Unfortunately, identifying appropriate models for the complete analysis and investigation of soft robots for design and control purposes can be problematic. This paper seeks to address this challenge by proposing two complementary modeling techniques for a particular type of soft robotic actuator known as a Fiber-Reinforced Elastomeric Enclosure (FREE). We propose that researchers can leverage multiple models to fill gaps in the understanding of the behavior of soft robots. We present and evaluate both a dynamic, lumped-parameter model and a finite element model to extend understanding of the practicability of FREEs in soft robotic applications. The results with the lumped-parameter model demonstrate that it predicts the actual rotational motion of a FREE with at most 4% error when a closed-loop controller is embedded in the system. Additionally, finite element analysis was used to study FREE design parameters as well as the workspace achieved with a module comprised of multiple FREEs. Our finite element results indicate that variations in the material properties of the elastic enclosure of a FREE are more significant than variations in fiber properties. Finally, finite element results show that a 30-degree difference in winding angle dramatically alters the shape of the workspace generated by four FREEs assembled into a module. Concludingly, comments are made about the relative advantages and limitations of lumped-parameter and finite element models of FREEs and FREE modules in providing useful insights into their behavior.

Robots can learn from humans by asking questions. In these questions the robot demonstrates a few different behaviors and asks the human for their favorite. But how should robots choose which questions to ask? Today's robots optimize for informative questions that actively probe the human's preferences as efficiently as possible. But while informative questions make sense from the robot's perspective, human onlookers often find them arbitrary and misleading. In this paper we formalize active preference-based learning from the human's perspective. We hypothesize that -- from the human's point-of-view -- the robot's questions reveal what the robot has and has not learned. Our insight enables robots to use questions to make their learning process transparent to the human operator. We develop and test a model that robots can leverage to relate the questions they ask to the information these questions reveal. We then introduce a trade-off between informative and revealing questions that considers both human and robot perspectives: a robot that optimizes for this trade-off actively gathers information from the human while simultaneously keeping the human up to date with what it has learned. We evaluate our approach across simulations, online surveys, and in-person user studies. Videos of our user studies and results are available here: https://youtu.be/tC6y_jHN7Vw.

The multidisciplinary nature of response robotics has brought about a diversified research community with extended expertise. Motivated by the recent accelerated rate of publications in the field, this paper analyzes the technical content, statistics, and implications of the literature from bibliometric standpoints. The aim is to study the global progress of response robotics research and identify the contemporary trends. To that end, we investigated the collaboration mapping together with the citation network to formally recognize impactful and contributing authors, publications, sources, institutions, funding agencies, and countries. We found how natural and human-made disasters contributed to forming productive regional research communities, while there are communities that only view response robotics as an application of their research. Furthermore, through an extensive discussion on the bibliometric results, we elucidated the philosophy behind research priority shifts in response robotics and presented our deliberations on future research directions.

In this paper, we propose a conditional early exiting framework for efficient video recognition. While existing works focus on selecting a subset of salient frames to reduce the computation costs, we propose to use a simple sampling strategy combined with conditional early exiting to enable efficient recognition. Our model automatically learns to process fewer frames for simpler videos and more frames for complex ones. To achieve this, we employ a cascade of gating modules to automatically determine the earliest point in processing where an inference is sufficiently reliable. We generate on-the-fly supervision signals to the gates to provide a dynamic trade-off between accuracy and computational cost. Our proposed model outperforms competing methods on three large-scale video benchmarks. In particular, on ActivityNet1.3 and mini-kinetics, we outperform the state-of-the-art efficient video recognition methods with 1.3$\times$ and 2.1$\times$ less GFLOPs, respectively. Additionally, our method sets a new state of the art for efficient video understanding on the HVU benchmark., Comment: CVPR 2021 | Oral paper

We propose Skip-Convolutions to leverage the large amount of redundancies in video streams and save computations. Each video is represented as a series of changes across frames and network activations, denoted as residuals. We reformulate standard convolution to be efficiently computed on residual frames: each layer is coupled with a binary gate deciding whether a residual is important to the model prediction,~\eg foreground regions, or it can be safely skipped, e.g. background regions. These gates can either be implemented as an efficient network trained jointly with convolution kernels, or can simply skip the residuals based on their magnitude. Gating functions can also incorporate block-wise sparsity structures, as required for efficient implementation on hardware platforms. By replacing all convolutions with Skip-Convolutions in two state-of-the-art architectures, namely EfficientDet and HRNet, we reduce their computational cost consistently by a factor of 3~4x for two different tasks, without any accuracy drop. Extensive comparisons with existing model compression, as well as image and video efficiency methods demonstrate that Skip-Convolutions set a new state-of-the-art by effectively exploiting the temporal redundancies in videos., Comment: CVPR 2021

A beat-to-beat Tele-fetal Monitoring and comparison with clinical data are studied with a wavelet transformation approach. Tele-fetal monitoring is a big progress toward a wearable medical device for a pregnant woman capable of obtaining prenatal care at home. We apply a wavelet transformation algorithm for fetal cardiac monitoring using a portable fetal Doppler medical device. Choosing an appropriate mother wavelet, 85 different mother wavelets are investigated. The efficiency of the proposed method is evaluated using two data sets including public and clinical. From publicly available data on PhysioBank, and simultaneous clinical measurement, we prove that the comparison between obtained fetal heart rate by the algorithm and the baselines yields a promising accuracy beyond 95%. Finally, we conclude that the proposed algorithm would be a robust technique for any similar tele-fetal monitoring approach., Comment: 12 pages, 12 figures

Rescue robots are expected to carry out reconnaissance and dexterity operations in unknown environments comprising unstructured obstacles. Although a wide variety of designs and implementations have been presented within the field of rescue robotics, embedding all mobility, dexterity, and reconnaissance capabilities in a single robot remains a challenging problem. This paper explains the design and implementation of Karo, a mobile robot that exhibits a high degree of mobility at the side of maintaining required dexterity and exploration capabilities for urban search and rescue (USAR) missions. We first elicit the system requirements of a standard rescue robot from the frameworks of Rescue Robot League (RRL) of RoboCup and then, propose the conceptual design of Karo by drafting a locomotion and manipulation system. Considering that, this work presents comprehensive design processes along with detail mechanical design of the robot's platform and its 7-DOF manipulator. Further, we present the design and implementation of the command and control system by discussing the robot's power system, sensors, and hardware systems. In conjunction with this, we elucidate the way that Karo's software system and human-robot interface are implemented and employed. Furthermore, we undertake extensive evaluations of Karo's field performance to investigate whether the principal objective of this work has been satisfied. We demonstrate that Karo has effectively accomplished assigned standardized rescue operations by evaluating all aspects of its capabilities in both RRL's test suites and training suites of a fire department. Finally, the comprehensiveness of Karo's capabilities has been verified by drawing quantitative comparisons between Karo's performance and other leading robots participating in RRL.