PyTorchKR Papers

최신 논문

총 9,714편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2109.10504

KD-VLP: Improving End-to-End Vision-and-Language Pretraining with Object Knowledge Distillation

Self-supervised vision-and-language pretraining (VLP) aims to learn transferable multi-modal representations from large-scale image-text data and to achieve strong performances on a broad scope of vision-language tasks after finetuning. Previous mainstream VLP approaches typically adopt a two-step strategy relying on external object detectors to encode images in a multi-modal Transformer framework, which suffer from restrictive object concept space, limited image context and inefficient computation. In this paper, we propose an object-aware end-to-end VLP framework, which directly feeds image grid features from CNNs into the Transformer and learns the multi-modal representations jointly. More importantly, we propose to perform object knowledge distillation to facilitate learning cross-modal alignment at different semantic levels. To achieve that, we design two novel pretext tasks by taking object features and their semantic labels from external detectors as supervision: 1.) Object-guided masked vision modeling task focuses on enforcing object-aware representation learning in the multi-modal Transformer; 2.) Phrase-region alignment task aims to improve cross-modal alignment by utilizing the similarities between noun phrases and object labels in the linguistic space. Extensive experiments on a wide range of vision-language tasks demonstrate the efficacy of our proposed framework, and we achieve competitive or superior performances over the existing pretraining strategies.

arXiv:2109.10852공식 코드 · Papers with Code

Pix2seq: A Language Modeling Framework for Object Detection

We present Pix2Seq, a simple and generic framework for object detection. Unlike existing approaches that explicitly integrate prior knowledge about the task, we cast object detection as a language modeling task conditioned on the observed pixel inputs. Object descriptions (e.g., bounding boxes and class labels) are expressed as sequences of discrete tokens, and we train a neural network to perceive the image and generate the desired sequence. Our approach is based mainly on the intuition that if a neural network knows about where and what the objects are, we just need to teach it how to read them out. Beyond the use of task-specific data augmentations, our approach makes minimal assumptions about the task, yet it achieves competitive results on the challenging COCO dataset, compared to highly specialized and well optimized detection algorithms.

arXiv:2109.09903공식 코드 · Papers with Code

AirDOS: Dynamic SLAM benefits from Articulated Objects

Dynamic Object-aware SLAM (DOS) exploits object-level information to enable robust motion estimation in dynamic environments. Existing methods mainly focus on identifying and excluding dynamic objects from the optimization. In this paper, we show that feature-based visual SLAM systems can also benefit from the presence of dynamic articulated objects by taking advantage of two observations: (1) The 3D structure of each rigid part of articulated object remains consistent over time; (2) The points on the same rigid part follow the same motion. In particular, we present AirDOS, a dynamic object-aware system that introduces rigidity and motion constraints to model articulated objects. By jointly optimizing the camera pose, object motion, and the object 3D structure, we can rectify the camera pose estimation, preventing tracking loss, and generate 4D spatio-temporal maps for both dynamic objects and static scenes. Experiments show that our algorithm improves the robustness of visual SLAM algorithms in challenging crowded urban environments. To the best of our knowledge, AirDOS is the first dynamic object-aware SLAM system demonstrating that camera pose estimation can be improved by incorporating dynamic articulated objects.

arXiv:2109.10187

Oriented Object Detection in Aerial Images Based on Area Ratio of Parallelogram

Oriented object detection is a challenging task in aerial images since the objects in aerial images are displayed in arbitrary directions and are frequently densely packed. The mainstream detectors describe rotating objects using a five-parament or eight-parament representations, which suffer from representation ambiguity for orientated object definition. In this paper, we propose a novel representation method based on area ratio of parallelogram, called ARP. Specifically, ARP regresses the minimum bounding rectangle of the oriented object and three area ratios. Three area ratios include the area ratio of a directed object to the smallest circumscribed rectangle and two parallelograms to the minimum circumscribed rectangle. It simplifies offset learning and eliminates the issue of angular periodicity or label point sequences for oriented objects. To further remedy the confusion issue of nearly horizontal objects, the area ratio between the object and its minimal circumscribed rectangle is employed to guide the selection of horizontal or oriented detection for each object. Moreover, the rotated efficient Intersection over Union (R-EIoU) loss with horizontal bounding box and three area ratios are designed to optimize the bounding box regression for rotating objects. Experimental results on remote sensing datasets, including HRSC2016, DOTA, and UCAS-AOD, show that our method achieves superior detection performance than many state-of-the-art approaches.

arXiv:2109.10115

StereOBJ-1M: Large-scale Stereo Image Dataset for 6D Object Pose Estimation

We present a large-scale stereo RGB image object pose estimation dataset named the $\textbf{StereOBJ-1M}$ dataset. The dataset is designed to address challenging cases such as object transparency, translucency, and specular reflection, in addition to the common challenges of occlusion, symmetry, and variations in illumination and environments. In order to collect data of sufficient scale for modern deep learning models, we propose a novel method for efficiently annotating pose data in a multi-view fashion that allows data capturing in complex and flexible environments. Fully annotated with 6D object poses, our dataset contains over 393K frames and over 1.5M annotations of 18 objects recorded in 182 scenes constructed in 11 different environments. The 18 objects include 8 symmetric objects, 7 transparent objects, and 8 reflective objects. We benchmark two state-of-the-art pose estimation frameworks on StereOBJ-1M as baselines for future work. We also propose a novel object-level pose optimization method for computing 6D pose from keypoint predictions in multiple images. Project website: https://sites.google.com/view/stereobj-1m.

arXiv:2109.09627

Superquadric Object Representation for Optimization-based Semantic SLAM

Introducing semantically meaningful objects to visual Simultaneous Localization And Mapping (SLAM) has the potential to improve both the accuracy and reliability of pose estimates, especially in challenging scenarios with significant view-point and appearance changes. However, how semantic objects should be represented for an efficient inclusion in optimization-based SLAM frameworks is still an open question. Superquadrics(SQs) are an efficient and compact object representation, able to represent most common object types to a high degree, and typically retrieved from 3D point-cloud data. However, accurate 3D point-cloud data might not be available in all applications. Recent advancements in machine learning enabled robust object recognition and semantic mask measurements from camera images under many different appearance conditions. We propose a pipeline to leverage such semantic mask measurements to fit SQ parameters to multi-view camera observations using a multi-stage initialization and optimization procedure. We demonstrate the system's ability to retrieve randomly generated SQ parameters from multi-view mask observations in preliminary simulation experiments and evaluate different initialization stages and cost functions.

arXiv:2109.09512

Landmark Policy Optimization for Object Navigation Task

This work studies object goal navigation task, which involves navigating to the closest object related to the given semantic category in unseen environments. Recent works have shown significant achievements both in the end-to-end Reinforcement Learning approach and modular systems, but need a big step forward to be robust and optimal. We propose a hierarchical method that incorporates standard task formulation and additional area knowledge as landmarks, with a way to extract these landmarks. In a hierarchy, a low level consists of separately trained algorithms to the most intuitive skills, and a high level decides which skill is needed at this moment. With all proposed solutions, we achieve a 0.75 success rate in a realistic Habitat simulator. After a small stage of additional model training in a reconstructed virtual area at a simulator, we successfully confirmed our results in a real-world case.

arXiv:2109.09188

DeepPoint: A Deep Learning Model for 3D Reconstruction in Point Clouds via mmWave Radar

Recent research has shown that mmWave radar sensing is effective for object detection in low visibility environments, which makes it an ideal technique in autonomous navigation systems such as autonomous vehicles. However, due to the characteristics of radar signals such as sparsity, low resolution, specularity, and high noise, it is still quite challenging to reconstruct 3D object shapes via mmWave radar sensing. Built on our recent proposed 3DRIMR (3D Reconstruction and Imaging via mmWave Radar), we introduce in this paper DeepPoint, a deep learning model that generates 3D objects in point cloud format that significantly outperforms the original 3DRIMR design. The model adopts a conditional Generative Adversarial Network (GAN) based deep neural network architecture. It takes as input the 2D depth images of an object generated by 3DRIMR's Stage 1, and outputs smooth and dense 3D point clouds of the object. The model consists of a novel generator network that utilizes a sequence of DeepPoint blocks or layers to extract essential features of the union of multiple rough and sparse input point clouds of an object when observed from various viewpoints, given that those input point clouds may contain many incorrect points due to the imperfect generation process of 3DRIMR's Stage 1. The design of DeepPoint adopts a deep structure to capture the global features of input point clouds, and it relies on an optimally chosen number of DeepPoint blocks and skip connections to achieve performance improvement over the original 3DRIMR design. Our experiments have demonstrated that this model significantly outperforms the original 3DRIMR and other standard techniques in reconstructing 3D objects.

arXiv:2109.09163공식 코드 · Papers with Code

CaTGrasp: Learning Category-Level Task-Relevant Grasping in Clutter from Simulation

Task-relevant grasping is critical for industrial assembly, where downstream manipulation tasks constrain the set of valid grasps. Learning how to perform this task, however, is challenging, since task-relevant grasp labels are hard to define and annotate. There is also yet no consensus on proper representations for modeling or off-the-shelf tools for performing task-relevant grasps. This work proposes a framework to learn task-relevant grasping for industrial objects without the need of time-consuming real-world data collection or manual annotation. To achieve this, the entire framework is trained solely in simulation, including supervised training with synthetic label generation and self-supervised, hand-object interaction. In the context of this framework, this paper proposes a novel, object-centric canonical representation at the category level, which allows establishing dense correspondence across object instances and transferring task-relevant grasps to novel instances. Extensive experiments on task-relevant grasping of densely-cluttered industrial objects are conducted in both simulation and real-world setups, demonstrating the effectiveness of the proposed framework. Code and data are available at https://sites.google.com/view/catgrasp.

arXiv:2109.09052

Object Tracking by Jointly Exploiting Frame and Event Domain

Inspired by the complementarity between conventional frame-based and bio-inspired event-based cameras, we propose a multi-modal based approach to fuse visual cues from the frame- and event-domain to enhance the single object tracking performance, especially in degraded conditions (e.g., scenes with high dynamic range, low light, and fast-motion objects). The proposed approach can effectively and adaptively combine meaningful information from both domains. Our approach's effectiveness is enforced by a novel designed cross-domain attention schemes, which can effectively enhance features based on self- and cross-domain attention schemes; The adaptiveness is guarded by a specially designed weighting scheme, which can adaptively balance the contribution of the two domains. To exploit event-based visual cues in single-object tracking, we construct a large-scale frame-event-based dataset, which we subsequently employ to train a novel frame-event fusion based model. Extensive experiments show that the proposed approach outperforms state-of-the-art frame-based tracking methods by at least 10.4% and 11.9% in terms of representative success rate and precision rate, respectively. Besides, the effectiveness of each key component of our approach is evidenced by our thorough ablation study.

arXiv:2109.09033공식 코드 · Papers with Code

Joint Distribution Alignment via Adversarial Learning for Domain Adaptive Object Detection

Unsupervised domain adaptive object detection aims to adapt a well-trained detector from its original source domain with rich labeled data to a new target domain with unlabeled data. Recently, mainstream approaches perform this task through adversarial learning, yet still suffer from two limitations. First, they mainly align marginal distribution by unsupervised cross-domain feature matching, and ignore each feature's categorical and positional information that can be exploited for conditional alignment; Second, they treat all classes as equally important for transferring cross-domain knowledge and ignore that different classes usually have different transferability. In this paper, we propose a joint adaptive detection framework (JADF) to address the above challenges. First, an end-to-end joint adversarial adaptation framework for object detection is proposed, which aligns both marginal and conditional distributions between domains without introducing any extra hyperparameter. Next, to consider the transferability of each object class, a metric for class-wise transferability assessment is proposed, which is incorporated into the JADF objective for domain adaptation. Further, an extended study from unsupervised domain adaptation (UDA) to unsupervised few-shot domain adaptation (UFDA) is conducted, where only a few unlabeled training images are available in unlabeled target domain. Extensive experiments validate that JADF is effective in both the UDA and UFDA settings, achieving significant performance gains over existing state-of-the-art cross-domain detection methods.

arXiv:2109.08973공식 코드 · Papers with Code

Hierarchical Policy for Non-prehensile Multi-object Rearrangement with Deep Reinforcement Learning and Monte Carlo Tree Search

Non-prehensile multi-object rearrangement is a robotic task of planning feasible paths and transferring multiple objects to their predefined target poses without grasping. It needs to consider how each object reaches the target and the order of object movement, which significantly deepens the complexity of the problem. To address these challenges, we propose a hierarchical policy to divide and conquer for non-prehensile multi-object rearrangement. In the high-level policy, guided by a designed policy network, the Monte Carlo Tree Search efficiently searches for the optimal rearrangement sequence among multiple objects, which benefits from imitation and reinforcement. In the low-level policy, the robot plans the paths according to the order of path primitives and manipulates the objects to approach the goal poses one by one. We verify through experiments that the proposed method can achieve a higher success rate, fewer steps, and shorter path length compared with the state-of-the-art.

arXiv:2109.08817

Learning to Regrasp by Learning to Place

In this paper, we explore whether a robot can learn to regrasp a diverse set of objects to achieve various desired grasp poses. Regrasping is needed whenever a robot's current grasp pose fails to perform desired manipulation tasks. Endowing robots with such an ability has applications in many domains such as manufacturing or domestic services. Yet, it is a challenging task due to the large diversity of geometry in everyday objects and the high dimensionality of the state and action space. In this paper, we propose a system for robots to take partial point clouds of an object and the supporting environment as inputs and output a sequence of pick-and-place operations to transform an initial object grasp pose to the desired object grasp poses. The key technique includes a neural stable placement predictor and a regrasp graph-based solution through leveraging and changing the surrounding environment. We introduce a new and challenging synthetic dataset for learning and evaluating the proposed approach. We demonstrate the effectiveness of our proposed system with both simulator and real-world experiments. More videos and visualization examples are available on our project webpage.

arXiv:2109.07991

ObjectFolder: A Dataset of Objects with Implicit Visual, Auditory, and Tactile Representations

Multisensory object-centric perception, reasoning, and interaction have been a key research topic in recent years. However, the progress in these directions is limited by the small set of objects available -- synthetic objects are not realistic enough and are mostly centered around geometry, while real object datasets such as YCB are often practically challenging and unstable to acquire due to international shipping, inventory, and financial cost. We present ObjectFolder, a dataset of 100 virtualized objects that addresses both challenges with two key innovations. First, ObjectFolder encodes the visual, auditory, and tactile sensory data for all objects, enabling a number of multisensory object recognition tasks, beyond existing datasets that focus purely on object geometry. Second, ObjectFolder employs a uniform, object-centric, and implicit representation for each object's visual textures, acoustic simulations, and tactile readings, making the dataset flexible to use and easy to share. We demonstrate the usefulness of our dataset as a testbed for multisensory perception and control by evaluating it on a variety of benchmark tasks, including instance recognition, cross-sensory retrieval, 3D reconstruction, and robotic grasping.

arXiv:2109.07843

Label Assignment Distillation for Object Detection

Knowledge distillation methods are proved to be promising in improving the performance of neural networks and no additional computational expenses are required during the inference time. For the sake of boosting the accuracy of object detection, a great number of knowledge distillation methods have been proposed particularly designed for object detection. However, most of these methods only focus on feature-level distillation and label-level distillation, leaving the label assignment step, a unique and paramount procedure for object detection, by the wayside. In this work, we come up with a simple but effective knowledge distillation approach focusing on label assignment in object detection, in which the positive and negative samples of student network are selected in accordance with the predictions of teacher network. Our method shows encouraging results on the MSCOCO2017 benchmark, and can not only be applied to both one-stage detectors and two-stage detectors but also be utilized orthogonally with other knowledge distillation methods.

arXiv:2109.07734

Few-Shot Object Detection by Attending to Per-Sample-Prototype

Few-shot object detection aims to detect instances of specific categories in a query image with only a handful of support samples. Although this takes less effort than obtaining enough annotated images for supervised object detection, it results in a far inferior performance compared to the conventional object detection methods. In this paper, we propose a meta-learning-based approach that considers the unique characteristics of each support sample. Rather than simply averaging the information of the support samples to generate a single prototype per category, our method can better utilize the information of each support sample by treating each support sample as an individual prototype. Specifically, we introduce two types of attention mechanisms for aggregating the query and support feature maps. The first is to refine the information of few-shot samples by extracting shared information between the support samples through attention. Second, each support sample is used as a class code to leverage the information by comparing similarities between each support feature and query features. Our proposed method is complementary to the previous methods, making it easy to plug and play for further improvement. We have evaluated our method on PASCAL VOC and COCO benchmarks, and the results verify the effectiveness of our method. In particular, the advantages of our method are maximized when there is more diversity among support data.

arXiv:2109.07559

Hybrid ICP

ICP algorithms typically involve a fixed choice of data association method and a fixed choice of error metric. In this paper, we propose Hybrid ICP, a novel and flexible ICP variant which dynamically optimises both the data association method and error metric based on the live image of an object and the current ICP estimate. We show that when used for object pose estimation, Hybrid ICP is more accurate and more robust to noise than other commonly used ICP variants. We also consider the setting where ICP is applied sequentially with a moving camera, and we study the trade-off between the accuracy of each ICP estimate and the number of ICP estimates available within a fixed amount of time.

arXiv:2109.07301

What Vision-Language Models `See' when they See Scenes

Images can be described in terms of the objects they contain, or in terms of the types of scene or place that they instantiate. In this paper we address to what extent pretrained Vision and Language models can learn to align descriptions of both types with images. We compare 3 state-of-the-art models, VisualBERT, LXMERT and CLIP. We find that (i) V&L models are susceptible to stylistic biases acquired during pretraining; (ii) only CLIP performs consistently well on both object- and scene-level descriptions. A follow-up ablation study shows that CLIP uses object-level information in the visual modality to align with scene-level textual descriptions.

arXiv:2109.07298공식 코드 · Papers with Code

FFAVOD: Feature Fusion Architecture for Video Object Detection

A significant amount of redundancy exists between consecutive frames of a video. Object detectors typically produce detections for one image at a time, without any capabilities for taking advantage of this redundancy. Meanwhile, many applications for object detection work with videos, including intelligent transportation systems, advanced driver assistance systems and video surveillance. Our work aims at taking advantage of the similarity between video frames to produce better detections. We propose FFAVOD, standing for feature fusion architecture for video object detection. We first introduce a novel video object detection architecture that allows a network to share feature maps between nearby frames. Second, we propose a feature fusion module that learns to merge feature maps to enhance them. We show that using the proposed architecture and the fusion module can improve the performance of three base object detectors on two object detection benchmarks containing sequences of moving road users. Additionally, to further increase performance, we propose an improvement to the SpotNet attention module. Using our architecture on the improved SpotNet detector, we obtain the state-of-the-art performance on the UA-DETRAC public benchmark as well as on the UAVDT dataset. Code is available at https://github.com/hu64/FFAVOD.

arXiv:2109.07107공식 코드 · Papers with Code

Anchor DETR: Query Design for Transformer-Based Object Detection

In this paper, we propose a novel query design for the transformer-based object detection. In previous transformer-based detectors, the object queries are a set of learned embeddings. However, each learned embedding does not have an explicit physical meaning and we cannot explain where it will focus on. It is difficult to optimize as the prediction slot of each object query does not have a specific mode. In other words, each object query will not focus on a specific region. To solved these problems, in our query design, object queries are based on anchor points, which are widely used in CNN-based detectors. So each object query focuses on the objects near the anchor point. Moreover, our query design can predict multiple objects at one position to solve the difficulty: "one region, multiple objects". In addition, we design an attention variant, which can reduce the memory cost while achieving similar or better performance than the standard attention in DETR. Thanks to the query design and the attention variant, the proposed detector that we called Anchor DETR, can achieve better performance and run faster than the DETR with 10$\times$ fewer training epochs. For example, it achieves 44.2 AP with 19 FPS on the MSCOCO dataset when using the ResNet50-DC5 feature for training 50 epochs. Extensive experiments on the MSCOCO benchmark prove the effectiveness of the proposed methods. Code is available at \url{https://github.com/megvii-research/AnchorDETR}.