Digital human motion synthesis is a vibrant research field with applications in movies, AR/VR, and video games. Whereas methods were proposed to generate natural and realistic human motions, most only focus on modeling humans and largely ignore object movements. Generating task-oriented human-object interaction motions in simulation is challenging. For different intents of using the objects, humans conduct various motions, which requires the human first to approach the objects and then make them move consistently with the human instead of staying still. Also, to deploy in downstream applications, the synthesized motions are desired to be flexible in length, providing options to personalize the predicted motions for various purposes. To this end, we propose TOHO: Task-Oriented Human-Object Interactions Generation with Implicit Neural Representations, which generates full human-object interaction motions to conduct specific tasks, given only the task type, the object, and a starting human status. TOHO generates human-object motions in three steps: 1) it first estimates the keyframe poses of conducting a task given the task type and object information; 2) then, it infills the keyframes and generates continuous motions; 3) finally, it applies a compact closed-form object motion estimation to generate the object motion. Our method generates continuous motions that are parameterized only by the temporal coordinate, which allows for upsampling or downsampling of the sequence to arbitrary frames and adjusting the motion speeds by designing the temporal coordinate vector. We demonstrate the effectiveness of our method, both qualitatively and quantitatively. This work takes a step further toward general human-scene interaction simulation.
Semantic image editing provides users with a flexible tool to modify a given image guided by a corresponding segmentation map. In this task, the features of the foreground objects and the backgrounds are quite different. However, all previous methods handle backgrounds and objects as a whole using a monolithic model. Consequently, they remain limited in processing content-rich images and suffer from generating unrealistic objects and texture-inconsistent backgrounds. To address this issue, we propose a novel paradigm, \textbf{S}emantic \textbf{I}mage \textbf{E}diting by \textbf{D}isentangling \textbf{O}bject and \textbf{B}ackground (\textbf{SIEDOB}), the core idea of which is to explicitly leverages several heterogeneous subnetworks for objects and backgrounds. First, SIEDOB disassembles the edited input into background regions and instance-level objects. Then, we feed them into the dedicated generators. Finally, all synthesized parts are embedded in their original locations and utilize a fusion network to obtain a harmonized result. Moreover, to produce high-quality edited images, we propose some innovative designs, including Semantic-Aware Self-Propagation Module, Boundary-Anchored Patch Discriminator, and Style-Diversity Object Generator, and integrate them into SIEDOB. We conduct extensive experiments on Cityscapes and ADE20K-Room datasets and exhibit that our method remarkably outperforms the baselines, especially in synthesizing realistic and diverse objects and texture-consistent backgrounds.
Recent open-vocabulary detection methods aim to detect novel objects by distilling knowledge from vision-language models (VLMs) trained on a vast amount of image-text pairs. To improve the effectiveness of these methods, researchers have utilized datasets with a large vocabulary that contains a large number of object classes, under the assumption that such data will enable models to extract comprehensive knowledge on the relationships between various objects and better generalize to unseen object classes. In this study, we argue that more fine-grained labels are necessary to extract richer knowledge about novel objects, including object attributes and relationships, in addition to their names. To address this challenge, we propose a simple and effective method named Pseudo Caption Labeling (PCL), which utilizes an image captioning model to generate captions that describe object instances from diverse perspectives. The resulting pseudo caption labels offer dense samples for knowledge distillation. On the LVIS benchmark, our best model trained on the de-duplicated VisualGenome dataset achieves an AP of 34.5 and an APr of 30.6, comparable to the state-of-the-art performance. PCL's simplicity and flexibility are other notable features, as it is a straightforward pre-processing technique that can be used with any image captioning model without imposing any restrictions on model architecture or training process.
We study the problem of object retrieval in scenarios where visual sensing is absent, object shapes are unknown beforehand and objects can move freely, like grabbing objects out of a drawer. Successful solutions require localizing free objects, identifying specific object instances, and then grasping the identified objects, only using touch feedback. Unlike vision, where cameras can observe the entire scene, touch sensors are local and only observe parts of the scene that are in contact with the manipulator. Moreover, information gathering via touch sensors necessitates applying forces on the touched surface which may disturb the scene itself. Reasoning with touch, therefore, requires careful exploration and integration of information over time -- a challenge we tackle. We present a system capable of using sparse tactile feedback from fingertip touch sensors on a dexterous hand to localize, identify and grasp novel objects without any visual feedback. Videos are available at https://taochenshh.github.io/projects/tactofind.
Real-world manipulation problems in heavy clutter require robots to reason about potential contacts with objects in the environment. We focus on pick-and-place style tasks to retrieve a target object from a shelf where some `movable' objects must be rearranged in order to solve the task. In particular, our motivation is to allow the robot to reason over and consider non-prehensile rearrangement actions that lead to complex robot-object and object-object interactions where multiple objects might be moved by the robot simultaneously, and objects might tilt, lean on each other, or topple. To support this, we query a physics-based simulator to forward simulate these interaction dynamics which makes action evaluation during planning computationally very expensive. To make the planner tractable, we establish a connection between the domain of Manipulation Among Movable Objects and Multi-Agent Pathfinding that lets us decompose the problem into two phases our M4M algorithm iterates over. First we solve a multi-agent planning problem that reasons about the configurations of movable objects but does not forward simulate a physics model. Next, an arm motion planning problem is solved that uses a physics-based simulator but does not search over possible configurations of movable objects. We run simulated and real-world experiments with the PR2 robot and compare against relevant baseline algorithms. Our results highlight that M4M generates complex 3D interactions, and solves at least twice as many problems as the baselines with competitive performance.
The ability to manipulate objects in a desired configurations is a fundamental requirement for robots to complete various practical applications. While certain goals can be achieved by picking and placing the objects of interest directly, object reorientation is needed for precise placement in most of the tasks. In such scenarios, the object must be reoriented and re-positioned into intermediate poses that facilitate accurate placement at the target pose. To this end, we propose a reorientation planning method, ReorientDiff, that utilizes a diffusion model-based approach. The proposed method employs both visual inputs from the scene, and goal-specific language prompts to plan intermediate reorientation poses. Specifically, the scene and language-task information are mapped into a joint scene-task representation feature space, which is subsequently leveraged to condition the diffusion model. The diffusion model samples intermediate poses based on the representation using classifier-free guidance and then uses gradients of learned feasibility-score models for implicit iterative pose-refinement. The proposed method is evaluated using a set of YCB-objects and a suction gripper, demonstrating a success rate of 95.2% in simulation. Overall, our study presents a promising approach to address the reorientation challenge in manipulation by learning a conditional distribution, which is an effective way to move towards more generalizable object manipulation. For more results, checkout our website: https://utkarshmishra04.github.io/ReorientDiff.
In-hand object manipulation is challenging to simulate due to complex contact dynamics, non-repetitive finger gaits, and the need to indirectly control unactuated objects. Further adapting a successful manipulation skill to new objects with different shapes and physical properties is a similarly challenging problem. In this work, we show that natural and robust in-hand manipulation of simple objects in a dynamic simulation can be learned from a high quality motion capture example via deep reinforcement learning with careful designs of the imitation learning problem. We apply our approach on both single-handed and two-handed dexterous manipulations of diverse object shapes and motions. We then demonstrate further adaptation of the example motion to a more complex shape through curriculum learning on intermediate shapes morphed between the source and target object. While a naive curriculum of progressive morphs often falls short, we propose a simple greedy curriculum search algorithm that can successfully apply to a range of objects such as a teapot, bunny, bottle, train, and elephant.
Finding localized correspondences across different images of the same object is crucial to understand its geometry. In recent years, this problem has seen remarkable progress with the advent of deep learning-based local image features and learnable matchers. Still, learnable matchers often underperform when there exists only small regions of co-visibility between image pairs (i.e. wide camera baselines). To address this problem, we leverage recent progress in coarse single-view geometry estimation methods. We propose LFM-3D, a Learnable Feature Matching framework that uses models based on graph neural networks and enhances their capabilities by integrating noisy, estimated 3D signals to boost correspondence estimation. When integrating 3D signals into the matcher model, we show that a suitable positional encoding is critical to effectively make use of the low-dimensional 3D information. We experiment with two different 3D signals - normalized object coordinates and monocular depth estimates - and evaluate our method on large-scale (synthetic and real) datasets containing object-centric image pairs across wide baselines. We observe strong feature matching improvements compared to 2D-only methods, with up to +6% total recall and +28% precision at fixed recall. Additionally, we demonstrate that the resulting improved correspondences lead to much higher relative posing accuracy for in-the-wild image pairs - up to 8.6% compared to the 2D-only approach.
One-to-one label assignment in object detection has successfully obviated the need for non-maximum suppression (NMS) as postprocessing and makes the pipeline end-to-end. However, it triggers a new dilemma as the widely used sparse queries cannot guarantee a high recall, while dense queries inevitably bring more similar queries and encounter optimization difficulties. As both sparse and dense queries are problematic, then what are the expected queries in end-to-end object detection? This paper shows that the solution should be Dense Distinct Queries (DDQ). Concretely, we first lay dense queries like traditional detectors and then select distinct ones for one-to-one assignments. DDQ blends the advantages of traditional and recent end-to-end detectors and significantly improves the performance of various detectors including FCN, R-CNN, and DETRs. Most impressively, DDQ-DETR achieves 52.1 AP on MS-COCO dataset within 12 epochs using a ResNet-50 backbone, outperforming all existing detectors in the same setting. DDQ also shares the benefit of end-to-end detectors in crowded scenes and achieves 93.8 AP on CrowdHuman. We hope DDQ can inspire researchers to consider the complementarity between traditional methods and end-to-end detectors. The source code can be found at \url{https://github.com/jshilong/DDQ}.
Object detection is one of the most important and fundamental aspects of computer vision tasks, which has been broadly utilized in pose estimation, object tracking and instance segmentation models. To obtain training data for object detection model efficiently, many datasets opt to obtain their unannotated data in video format and the annotator needs to draw a bounding box around each object in the images. Annotating every frame from a video is costly and inefficient since many frames contain very similar information for the model to learn from. How to select the most informative frames from a video to annotate has become a highly practical task to solve but attracted little attention in research. In this paper, we proposed a novel active learning algorithm for object detection models to tackle this problem. In the proposed active learning algorithm, both classification and localization informativeness of unlabelled data are measured and aggregated. Utilizing the temporal information from video frames, two novel localization informativeness measurements are proposed. Furthermore, a weight curve is proposed to avoid querying adjacent frames. Proposed active learning algorithm with multiple configurations was evaluated on the MuPoTS dataset and FootballPD dataset.
In autonomous driving, data augmentation is commonly used for improving 3D object detection. The most basic methods include insertion of copied objects and rotation and scaling of the entire training frame. Numerous variants have been developed as well. The existing methods, however, are considerably limited when compared to the variety of the real world possibilities. In this work, we develop a diversified and realistic augmentation method that can flexibly construct a whole-body object, freely locate and rotate the object, and apply self-occlusion and external-occlusion accordingly. To improve the diversity of the whole-body object construction, we develop an iterative method that stochastically combines multiple objects observed from the real world into a single object. Unlike the existing augmentation methods, the constructed objects can be randomly located and rotated in the training frame because proper occlusions can be reflected to the whole-body objects in the final step. Finally, proper self-occlusion at each local object level and external-occlusion at the global frame level are applied using the Hidden Point Removal (HPR) algorithm that is computationally efficient. HPR is also used for adaptively controlling the point density of each object according to the object's distance from the LiDAR. Experiment results show that the proposed DR.CPO algorithm is data-efficient and model-agnostic without incurring any computational overhead. Also, DR.CPO can improve mAP performance by 2.08% when compared to the best 3D detection result known for KITTI dataset. The code is available at https://github.com/SNU-DRL/DRCPO.git
A key challenge for LiDAR-based 3D object detection is to capture sufficient features from large scale 3D scenes especially for distant or/and occluded objects. Albeit recent efforts made by Transformers with the long sequence modeling capability, they fail to properly balance the accuracy and efficiency, suffering from inadequate receptive fields or coarse-grained holistic correlations. In this paper, we propose an Octree-based Transformer, named OcTr, to address this issue. It first constructs a dynamic octree on the hierarchical feature pyramid through conducting self-attention on the top level and then recursively propagates to the level below restricted by the octants, which captures rich global context in a coarse-to-fine manner while maintaining the computational complexity under control. Furthermore, for enhanced foreground perception, we propose a hybrid positional embedding, composed of the semantic-aware positional embedding and attention mask, to fully exploit semantic and geometry clues. Extensive experiments are conducted on the Waymo Open Dataset and KITTI Dataset, and OcTr reaches newly state-of-the-art results.
Recent successes in image synthesis are powered by large-scale diffusion models. However, most methods are currently limited to either text- or image-conditioned generation for synthesizing an entire image, texture transfer or inserting objects into a user-specified region. In contrast, in this work we focus on synthesizing complex interactions (ie, an articulated hand) with a given object. Given an RGB image of an object, we aim to hallucinate plausible images of a human hand interacting with it. We propose a two-step generative approach: a LayoutNet that samples an articulation-agnostic hand-object-interaction layout, and a ContentNet that synthesizes images of a hand grasping the object given the predicted layout. Both are built on top of a large-scale pretrained diffusion model to make use of its latent representation. Compared to baselines, the proposed method is shown to generalize better to novel objects and perform surprisingly well on out-of-distribution in-the-wild scenes of portable-sized objects. The resulting system allows us to predict descriptive affordance information, such as hand articulation and approaching orientation. Project page: https://judyye.github.io/affordiffusion-www
Most recent 6D object pose estimation methods first use object detection to obtain 2D bounding boxes before actually regressing the pose. However, the general object detection methods they use are ill-suited to handle cluttered scenes, thus producing poor initialization to the subsequent pose network. To address this, we propose a rigidity-aware detection method exploiting the fact that, in 6D pose estimation, the target objects are rigid. This lets us introduce an approach to sampling positive object regions from the entire visible object area during training, instead of naively drawing samples from the bounding box center where the object might be occluded. As such, every visible object part can contribute to the final bounding box prediction, yielding better detection robustness. Key to the success of our approach is a visibility map, which we propose to build using a minimum barrier distance between every pixel in the bounding box and the box boundary. Our results on seven challenging 6D pose estimation datasets evidence that our method outperforms general detection frameworks by a large margin. Furthermore, combined with a pose regression network, we obtain state-of-the-art pose estimation results on the challenging BOP benchmark.
We propose a new setting for detecting unseen objects called Zero-shot Annotation object Detection (ZAD). It expands the zero-shot object detection setting by allowing the novel objects to exist in the training images and restricts the additional information the detector uses to novel category names. Recently, to detect unseen objects, large-scale vision-language models (e.g., CLIP) are leveraged by different methods. The distillation-based methods have good overall performance but suffer from a long training schedule caused by two factors. First, existing work creates distillation regions biased to the base categories, which limits the distillation of novel category information. Second, directly using the raw feature from CLIP for distillation neglects the domain gap between the training data of CLIP and the detection datasets, which makes it difficult to learn the mapping from the image region to the vision-language feature space. To solve these problems, we propose Efficient feature distillation for Zero-shot Annotation object Detection (EZAD). Firstly, EZAD adapts the CLIP's feature space to the target detection domain by re-normalizing CLIP; Secondly, EZAD uses CLIP to generate distillation proposals with potential novel category names to avoid the distillation being overly biased toward the base categories. Finally, EZAD takes advantage of semantic meaning for regression to further improve the model performance. As a result, EZAD outperforms the previous distillation-based methods in COCO by 4% with a much shorter training schedule and achieves a 3% improvement on the LVIS dataset. Our code is available at https://github.com/dragonlzm/EZAD
Human-Object Interaction (HOI) detection recognizes how persons interact with objects, which is advantageous in autonomous systems such as self-driving vehicles and collaborative robots. However, current HOI detectors are often plagued by model inefficiency and unreliability when making a prediction, which consequently limits its potential for real-world scenarios. In this paper, we address these challenges by proposing ERNet, an end-to-end trainable convolutional-transformer network for HOI detection. The proposed model employs an efficient multi-scale deformable attention to effectively capture vital HOI features. We also put forward a novel detection attention module to adaptively generate semantically rich instance and interaction tokens. These tokens undergo pre-emptive detections to produce initial region and vector proposals that also serve as queries which enhances the feature refinement process in the transformer decoders. Several impactful enhancements are also applied to improve the HOI representation learning. Additionally, we utilize a predictive uncertainty estimation framework in the instance and interaction classification heads to quantify the uncertainty behind each prediction. By doing so, we can accurately and reliably predict HOIs even under challenging scenarios. Experiment results on the HICO-Det, V-COCO, and HOI-A datasets demonstrate that the proposed model achieves state-of-the-art performance in detection accuracy and training efficiency. Codes are publicly available at https://github.com/Monash-CyPhi-AI-Research-Lab/ernet.
Object tracking (OT) aims to estimate the positions of target objects in a video sequence. Depending on whether the initial states of target objects are specified by provided annotations in the first frame or the categories, OT could be classified as instance tracking (e.g., SOT and VOS) and category tracking (e.g., MOT, MOTS, and VIS) tasks. Combing the advantages of the best practices developed in both communities, we propose a novel tracking-with-detection paradigm, where tracking supplements appearance priors for detection and detection provides tracking with candidate bounding boxes for association. Equipped with such a design, a unified tracking model, OmniTracker, is further presented to resolve all the tracking tasks with a fully shared network architecture, model weights, and inference pipeline. Extensive experiments on 7 tracking datasets, including LaSOT, TrackingNet, DAVIS16-17, MOT17, MOTS20, and YTVIS19, demonstrate that OmniTracker achieves on-par or even better results than both task-specific and unified tracking models.
Previous works on video object segmentation (VOS) are trained on densely annotated videos. Nevertheless, acquiring annotations in pixel level is expensive and time-consuming. In this work, we demonstrate the feasibility of training a satisfactory VOS model on sparsely annotated videos-we merely require two labeled frames per training video while the performance is sustained. We term this novel training paradigm as two-shot video object segmentation, or two-shot VOS for short. The underlying idea is to generate pseudo labels for unlabeled frames during training and to optimize the model on the combination of labeled and pseudo-labeled data. Our approach is extremely simple and can be applied to a majority of existing frameworks. We first pre-train a VOS model on sparsely annotated videos in a semi-supervised manner, with the first frame always being a labeled one. Then, we adopt the pre-trained VOS model to generate pseudo labels for all unlabeled frames, which are subsequently stored in a pseudo-label bank. Finally, we retrain a VOS model on both labeled and pseudo-labeled data without any restrictions on the first frame. For the first time, we present a general way to train VOS models on two-shot VOS datasets. By using 7.3% and 2.9% labeled data of YouTube-VOS and DAVIS benchmarks, our approach achieves comparable results in contrast to the counterparts trained on fully labeled set. Code and models are available at https://github.com/yk-pku/Two-shot-Video-Object-Segmentation.
We propose NEMTO, the first end-to-end neural rendering pipeline to model 3D transparent objects with complex geometry and unknown indices of refraction. Commonly used appearance modeling such as the Disney BSDF model cannot accurately address this challenging problem due to the complex light paths bending through refractions and the strong dependency of surface appearance on illumination. With 2D images of the transparent object as input, our method is capable of high-quality novel view and relighting synthesis. We leverage implicit Signed Distance Functions (SDF) to model the object geometry and propose a refraction-aware ray bending network to model the effects of light refraction within the object. Our ray bending network is more tolerant to geometric inaccuracies than traditional physically-based methods for rendering transparent objects. We provide extensive evaluations on both synthetic and real-world datasets to demonstrate our high-quality synthesis and the applicability of our method.
In this paper, we propose a long-sequence modeling framework, named StreamPETR, for multi-view 3D object detection. Built upon the sparse query design in the PETR series, we systematically develop an object-centric temporal mechanism. The model is performed in an online manner and the long-term historical information is propagated through object queries frame by frame. Besides, we introduce a motion-aware layer normalization to model the movement of the objects. StreamPETR achieves significant performance improvements only with negligible computation cost, compared to the single-frame baseline. On the standard nuScenes benchmark, it is the first online multi-view method that achieves comparable performance (67.6% NDS & 65.3% AMOTA) with lidar-based methods. The lightweight version realizes 45.0% mAP and 31.7 FPS, outperforming the state-of-the-art method (SOLOFusion) by 2.3% mAP and 1.8x faster FPS. Code has been available at https://github.com/exiawsh/StreamPETR.git.