arXiv subjects

Wei Jing

Publications and source records attributed to Wei Jing.

At least 37 records · Page 2Linked to original sources

arXiv2022

End-to-end Reinforcement Learning of Robotic Manipulation with Robust Keypoints Representation

We present an end-to-end Reinforcement Learning(RL) framework for robotic manipulation tasks, using a robust and efficient keypoints representation. The proposed method learns keypoints from camera images as the state representation, through a self-supervised autoencoder architecture. The keypoints encode the geometric information, as well as the relationship of the tool and target in a compact representation to ensure efficient and robust learning. After keypoints learning, the RL step then learns the robot motion from the extracted keypoints state representation. The keypoints and RL learning processes are entirely done in the simulated environment. We demonstrate the effectiveness of the proposed method on robotic manipulation tasks including grasping and pushing, in different scenarios. We also investigate the generalization capability of the trained model. In addition to the robust keypoints representation, we further apply domain randomization and adversarial training examples to achieve zero-shot sim-to-real transfer in real-world robotic manipulation tasks.

cs.RO↗

arXiv2022

Variance matrix priors for Dirichlet process mixture models with Gaussian kernels

The Dirichlet Process Mixture Model (DPMM) is a Bayesian non-parametric approach widely used for density estimation and clustering. In this manuscript, we study the choice of prior for the variance or precision matrix when Gaussian kernels are adopted. Typically, in the relevant literature, the assessment of mixture models is done by considering observations in a space of only a handful of dimensions. Instead, we are concerned with more realistic problems of higher dimensionality, in a space of up to 20 dimensions. We observe that the choice of prior is increasingly important as the dimensionality of the problem increases. After identifying certain undesirable properties of standard priors in problems of higher dimensionality, we review and implement possible alternative priors. The most promising priors are identified, as well as other factors that affect the convergence of MCMC samplers. Our results show that the choice of prior is critical for deriving reliable posterior inferences. This manuscript offers a thorough overview and comparative investigation into possible priors, with detailed guidelines for their implementation. Although our work focuses on the use of the DPMM in clustering, it is also applicable to density estimation.

stat.ME↗

arXiv2021

Towards Debiasing Temporal Sentence Grounding in Video

The temporal sentence grounding in video (TSGV) task is to locate a temporal moment from an untrimmed video, to match a language query, i.e., a sentence. Without considering bias in moment annotations (e.g., start and end positions in a video), many models tend to capture statistical regularities of the moment annotations, and do not well learn cross-modal reasoning between video and language query. In this paper, we propose two debiasing strategies, data debiasing and model debiasing, to "force" a TSGV model to capture cross-modal interactions. Data debiasing performs data oversampling through video truncation to balance moment temporal distribution in train set. Model debiasing leverages video-only and query-only models to capture the distribution bias, and forces the model to learn cross-modal interactions. Using VSLNet as the base model, we evaluate impact of the two strategies on two datasets that contain out-of-distribution test instances. Results show that both strategies are effective in improving model generalization capability. Equipped with both debiasing strategies, VSLNet achieves best results on both datasets.

cs.CV↗

arXiv2021

Domain Generalization for Vision-based Driving Trajectory Generation

One of the challenges in vision-based driving trajectory generation is dealing with out-of-distribution scenarios. In this paper, we propose a domain generalization method for vision-based driving trajectory generation for autonomous vehicles in urban environments, which can be seen as a solution to extend the Invariant Risk Minimization (IRM) method in complex problems. We leverage an adversarial learning approach to train a trajectory generator as the decoder. Based on the pre-trained decoder, we infer the latent variables corresponding to the trajectories, and pre-train the encoder by regressing the inferred latent variable. Finally, we fix the decoder but fine-tune the encoder with the final trajectory loss. We compare our proposed method with the state-of-the-art trajectory generation method and some recent domain generalization methods on both datasets and simulation, demonstrating that our method has better generalization ability.

cs.CV↗

arXiv2021

Parallel Attention Network with Sequence Matching for Video Grounding

Given a video, video grounding aims to retrieve a temporal moment that semantically corresponds to a language query. In this work, we propose a Parallel Attention Network with Sequence matching (SeqPAN) to address the challenges in this task: multi-modal representation learning, and target moment boundary prediction. We design a self-guided parallel attention module to effectively capture self-modal contexts and cross-modal attentive information between video and text. Inspired by sequence labeling tasks in natural language processing, we split the ground truth moment into begin, inside, and end regions. We then propose a sequence matching strategy to guide start/end boundary predictions using region labels. Experimental results on three datasets show that SeqPAN is superior to state-of-the-art methods. Furthermore, the effectiveness of the self-guided parallel attention module and the sequence matching module is verified.

cs.CL↗

arXiv2021

Video Corpus Moment Retrieval with Contrastive Learning

Given a collection of untrimmed and unsegmented videos, video corpus moment retrieval (VCMR) is to retrieve a temporal moment (i.e., a fraction of a video) that semantically corresponds to a given text query. As video and text are from two distinct feature spaces, there are two general approaches to address VCMR: (i) to separately encode each modality representations, then align the two modality representations for query processing, and (ii) to adopt fine-grained cross-modal interaction to learn multi-modal representations for query processing. While the second approach often leads to better retrieval accuracy, the first approach is far more efficient. In this paper, we propose a Retrieval and Localization Network with Contrastive Learning (ReLoCLNet) for VCMR. We adopt the first approach and introduce two contrastive learning objectives to refine video encoder and text encoder to learn video and text representations separately but with better alignment for VCMR. The video contrastive learning (VideoCL) is to maximize mutual information between query and candidate video at video-level. The frame contrastive learning (FrameCL) aims to highlight the moment region corresponds to the query at frame-level, within a video. Experimental results show that, although ReLoCLNet encodes text and video separately for efficiency, its retrieval accuracy is comparable with baselines adopting cross-modal interaction learning.

cs.CL↗

arXiv2021

6D Pose Estimation with Correlation Fusion

6D object pose estimation is widely applied in robotic tasks such as grasping and manipulation. Prior methods using RGB-only images are vulnerable to heavy occlusion and poor illumination, so it is important to complement them with depth information. However, existing methods using RGB-D data cannot adequately exploit consistent and complementary information between RGB and depth modalities. In this paper, we present a novel method to effectively consider the correlation within and across both modalities with attention mechanism to learn discriminative and compact multi-modal features. Then, effective fusion strategies for intra- and inter-correlation modules are explored to ensure efficient information flow between RGB and depth. To our best knowledge, this is the first work to explore effective intra- and inter-modality fusion in 6D pose estimation. The experimental results show that our method can achieve the state-of-the-art performance on LineMOD and YCB-Video dataset. We also demonstrate that the proposed method can benefit a real-world robot grasping task by providing accurate object pose estimation.

cs.CV↗

arXiv2021

Natural Language Video Localization: A Revisit in Span-based Question Answering Framework

Natural Language Video Localization (NLVL) aims to locate a target moment from an untrimmed video that semantically corresponds to a text query. Existing approaches mainly solve the NLVL problem from the perspective of computer vision by formulating it as ranking, anchor, or regression tasks. These methods suffer from large performance degradation when localizing on long videos. In this work, we address the NLVL from a new perspective, i.e., span-based question answering (QA), by treating the input video as a text passage. We propose a video span localizing network (VSLNet), on top of the standard span-based QA framework (named VSLBase), to address NLVL. VSLNet tackles the differences between NLVL and span-based QA through a simple yet effective query-guided highlighting (QGH) strategy. QGH guides VSLNet to search for the matching video span within a highlighted region. To address the performance degradation on long videos, we further extend VSLNet to VSLNet-L by applying a multi-scale split-and-concatenation strategy. VSLNet-L first splits the untrimmed video into short clip segments; then, it predicts which clip segment contains the target moment and suppresses the importance of other segments. Finally, the clip segments are concatenated, with different confidences, to locate the target moment accurately. Extensive experiments on three benchmark datasets show that the proposed VSLNet and VSLNet-L outperform the state-of-the-art methods; VSLNet-L addresses the issue of performance degradation on long videos. Our study suggests that the span-based QA framework is an effective strategy to solve the NLVL problem.

cs.CL↗

arXiv2020

Context Modeling with Evidence Filter for Multiple Choice Question Answering

Multiple-Choice Question Answering (MCQA) is a challenging task in machine reading comprehension. The main challenge in MCQA is to extract "evidence" from the given context that supports the correct answer. In the OpenbookQA dataset, the requirement of extracting "evidence" is particularly important due to the mutual independence of sentences in the context. Existing work tackles this problem by annotated evidence or distant supervision with rules which overly rely on human efforts. To address the challenge, we propose a simple yet effective approach termed evidence filtering to model the relationships between the encoded contexts with respect to different options collectively and to potentially highlight the evidence sentences and filter out unrelated sentences. In addition to the effective reduction of human efforts of our approach compared, through extensive experiments on OpenbookQA, we show that the proposed approach outperforms the models that use the same backbone and more training data; and our parameter analysis also demonstrates the interpretability of our approach.

cs.CL↗

arXiv2020

KOVIS: Keypoint-based Visual Servoing with Zero-Shot Sim-to-Real Transfer for Robotics Manipulation

We present KOVIS, a novel learning-based, calibration-free visual servoing method for fine robotic manipulation tasks with eye-in-hand stereo camera system. We train the deep neural network only in the simulated environment; and the trained model could be directly used for real-world visual servoing tasks. KOVIS consists of two networks. The first keypoint network learns the keypoint representation from the image using with an autoencoder. Then the visual servoing network learns the motion based on keypoints extracted from the camera image. The two networks are trained end-to-end in the simulated environment by self-supervised learning without manual data labeling. After training with data augmentation, domain randomization, and adversarial examples, we are able to achieve zero-shot sim-to-real transfer to real-world robotic manipulation tasks. We demonstrate the effectiveness of the proposed method in both simulated environment and real-world experiment with different robotic manipulation tasks, including grasping, peg-in-hole insertion with 4mm clearance, and M13 screw insertion. The demo video is available at http://youtu.be/gfBJBR2tDzA

cs.RO↗

arXiv2020

Multi-UAV Coverage Path Planning for the Inspection of Large and Complex Structures

We present a multi-UAV Coverage Path Planning (CPP) framework for the inspection of large-scale, complex 3D structures. In the proposed sampling-based coverage path planning method, we formulate the multi-UAV inspection applications as a multi-agent coverage path planning problem. By combining two NP-hard problems: Set Covering Problem (SCP) and Vehicle Routing Problem (VRP), a Set-Covering Vehicle Routing Problem (SC-VRP) is formulated and subsequently solved by a modified Biased Random Key Genetic Algorithm (BRKGA) with novel, efficient encoding strategies and local improvement heuristics. We test our proposed method for several complex 3D structures with the 3D model extracted from OpenStreetMap. The proposed method outperforms previous methods, by reducing the length of the planned inspection path by up to 48%

cs.RO↗

arXiv2020

Span-based Localizing Network for Natural Language Video Localization

Given an untrimmed video and a text query, natural language video localization (NLVL) is to locate a matching span from the video that semantically corresponds to the query. Existing solutions formulate NLVL either as a ranking task and apply multimodal matching architecture, or as a regression task to directly regress the target video span. In this work, we address NLVL task with a span-based QA approach by treating the input video as text passage. We propose a video span localizing network (VSLNet), on top of the standard span-based QA framework, to address NLVL. The proposed VSLNet tackles the differences between NLVL and span-based QA through a simple yet effective query-guided highlighting (QGH) strategy. The QGH guides VSLNet to search for matching video span within a highlighted region. Through extensive experiments on three benchmark datasets, we show that the proposed VSLNet outperforms the state-of-the-art methods; and adopting span-based QA framework is a promising direction to solve NLVL.

cs.CL↗

arXiv2020

Feature Super-Resolution Based Facial Expression Recognition for Multi-scale Low-Resolution Faces

Facial Expressions Recognition(FER) on low-resolution images is necessary for applications like group expression recognition in crowd scenarios(station, classroom etc.). Classifying a small size facial image into the right expression category is still a challenging task. The main cause of this problem is the loss of discriminative feature due to reduced resolution. Super-resolution method is often used to enhance low-resolution images, but the performance on FER task is limited when on images of very low resolution. In this work, inspired by feature super-resolution methods for object detection, we proposed a novel generative adversary network-based feature level super-resolution method for robust facial expression recognition(FSR-FER). In particular, a pre-trained FER model was employed as feature extractor, and a generator network G and a discriminator network D are trained with features extracted from images of low resolution and original high resolution. Generator network G tries to transform features of low-resolution images to more discriminative ones by making them closer to the ones of corresponding high-resolution images. For better classification performance, we also proposed an effective classification-aware loss re-weighting strategy based on the classification probability calculated by a fixed FER model to make our model focus more on samples that are easily misclassified. Experiment results on Real-World Affective Faces (RAF) Database demonstrate that our method achieves satisfying results on various down-sample factors with a single model and has better performance on low-resolution images compared with methods using image super-resolution and expression recognition separately.

cs.CV↗

arXiv2019

RoboCoDraw: Robotic Avatar Drawing with GAN-based Style Transfer and Time-efficient Path Optimization

Robotic drawing has become increasingly popular as an entertainment and interactive tool. In this paper we present RoboCoDraw, a real-time collaborative robot-based drawing system that draws stylized human face sketches interactively in front of human users, by using the Generative Adversarial Network (GAN)-based style transfer and a Random-Key Genetic Algorithm (RKGA)-based path optimization. The proposed RoboCoDraw system takes a real human face image as input, converts it to a stylized avatar, then draws it with a robotic arm. A core component in this system is the Avatar-GAN proposed by us, which generates a cartoon avatar face image from a real human face. AvatarGAN is trained with unpaired face and avatar images only and can generate avatar images of much better likeness with human face images in comparison with the vanilla CycleGAN. After the avatar image is generated, it is fed to a line extraction algorithm and converted to sketches. An RKGA-based path optimization algorithm is applied to find a time-efficient robotic drawing path to be executed by the robotic arm. We demonstrate the capability of RoboCoDraw on various face images using a lightweight, safe collaborative robot UR5.

cs.RO↗

arXiv2019

Efficient Robotic Task Generalization Using Deep Model Fusion Reinforcement Learning

Learning-based methods have been used to pro-gram robotic tasks in recent years. However, extensive training is usually required not only for the initial task learning but also for generalizing the learned model to the same task but in different environments. In this paper, we propose a novel Deep Reinforcement Learning algorithm for efficient task generalization and environment adaptation in the robotic task learning problem. The proposed method is able to efficiently generalize the previously learned task by model fusion to solve the environment adaptation problem. The proposed Deep Model Fusion (DMF) method reuses and combines the previously trained model to improve the learning efficiency and results.Besides, we also introduce a Multi-objective Guided Reward(MGR) shaping technique to further improve training efficiency.The proposed method was benchmarked with previous methods in various environments to validate its effectiveness.

cs.RO↗

arXiv2019

On the correspondence of deviances and maximum likelihood and interval estimates from log-linear to logistic regression modelling

Consider a set of categorical variables $\mathcal{P}$ where at least one, denoted by $Y$, is binary. The log-linear model that describes the counts in the resulting contingency table implies a specific logistic regression model, with the binary variable as the outcome. Extending results in Christensen (1997), by also considering the case where factors present in the contingency table disappear from the logistic regression model, we prove that the Maximum Likelihood Estimate (MLE) for the parameters of the logistic regression equals the MLE for the corresponding parameters of the log-linear model. We prove that, asymptotically, standard errors for the two sets of parameters are also equal. Subsequently, Wald confidence intervals are asymptotically equal. These results demonstrate the extent to which inferences from the log-linear framework can be translated to inferences within the logistic regression framework, on the magnitude of main effects and interactions. Finally, we prove that the deviance of the log-linear model is equal to the deviance of the corresponding logistic regression, provided that the latter is fitted to a dataset where no cell observations are merged when one or more factors in $\mathcal{P} \setminus \{ Y \}$ become obsolete. We illustrate the derived results with the analysis of a real dataset.

stat.ME↗

arXiv2019

Constrained Heterogeneous Vehicle Path Planning for Large-area Coverage

There is a strong demand for covering a large area autonomously by multiple UAVs (Unmanned Aerial Vehicles) supported by a ground vehicle. Limited by UAVs' battery life and communication distance, complete coverage of large areas typically involves multiple take-offs and landings to recharge batteries, and the transportation of UAVs between operation areas by a ground vehicle. In this paper, we introduce a novel large-area-coverage planning framework which collectively optimizes the paths for aerial and ground vehicles. Our method first partitions a large area into sub-areas, each of which a given fleet of UAVs can cover without recharging batteries. UAV operation routes, or trails, are then generated for each sub-area. Next, the assignment of trials to different UAVs and the order in which UAVs visit their assigned trails are simultaneously optimized to minimize the total UAV flight distance. Finally, a ground vehicle transportation path which visits all sub-areas is found by solving an asymmetric traveling salesman problem (ATSP). Although finding the globally optimal trail assignment and transition paths can be formulated as a Mixed Integer Quadratic Program (MIQP), the MIQP is intractable even for small problems. We show that the solution time can be reduced to close-to-real-time levels by first finding a feasible solution using a Random Key Genetic Algorithm (RKGA), which is then locally optimized by solving a much smaller MIQP.

cs.RO↗

arXiv2019

Coverage Path Planning using Path Primitive Sampling and Primitive Coverage Graph for Visual Inspection

Planning the path to gather the surface information of the target objects is crucial to improve the efficiency of and reduce the overall cost, for visual inspection applications with Unmanned Aerial Vehicles (UAVs). Coverage Path Planning (CPP) problem is often formulated for these inspection applications because of the coverage requirement. Traditionally, researchers usually plan and optimize the viewpoints to capture the surface information first, and then optimize the path to visit the selected viewpoints. In this paper, we propose a novel planning method to directly sample and plan the inspection path for a camera-equipped UAV to acquire visual and geometric information of the target structures as a video stream setting in complex 3D environment. The proposed planning method first generates via-points and path primitives around the target object by using sampling methods based on voxel dilation and subtraction. A novel Primitive Coverage Graph (PCG) is then proposed to encode the topological information, flying distances, and visibility information, with the sampled via-points and path primitives. Finally graph search is performed to find the resultant path in the PCG to complete the inspection task with the coverage requirements. The effectiveness of the proposed method is demonstrated through simulation and field tests in this paper.

cs.RO↗