Recently, three-dimensional (3D) detection based on stereo images has progressed remarkably; however, most advanced methods adopt anchor-based two-dimensional (2D) detection or depth estimation to address this problem. Nevertheless, high computational cost inhibits these methods from achieving real-time performance. In this study, we propose a 3D object detection method, Stereo CenterNet (SC), using geometric information in stereo imagery. SC predicts the four semantic key points of the 3D bounding box of the object in space and utilizes 2D left and right boxes, 3D dimension, orientation, and key points to restore the bounding box of the object in the 3D space. Subsequently, we adopt an improved photometric alignment module to further optimize the position of the 3D bounding box. Experiments conducted on the KITTI dataset indicate that the proposed SC exhibits the best speed-accuracy trade-off among advanced methods without using extra data.
StereoCarla: A High-Fidelity Driving Dataset for Generalizable Stereo
School of Computer Science, Wuhan University（武汉大学计算机学院）；IAIR, Xi’an Jiaotong University（西安交通大学IAIR）；Waytous；CASIA（中国科学院自动化所）；University of Bologna（博洛尼亚大学）；University of California, Berkeley（加州大学伯克利分校）；AIR, Tsinghua University（清华大学AIR）
Recently, there are emerging many stereo matching methods for autonomous driving based on unsupervised learning. Most of them take advantage of reconstruction losses to remove dependency on disparity groundtruth. Occlusion handling is a challenging problem in stereo matching, especially for unsupervised methods. Previous unsupervised methods failed to take full advantage of geometry properties in occlusion handling. In this paper, we introduce an effective way to detect occlusion regions and propose a novel unsupervised training strategy to deal with occlusion that only uses the predicted left…
Although the number of camera-based sensors mounted on vehicles has recently increased dramatically, robust and accurate object velocity detection is difficult. Additionally, it is still common to use radar as a fusion system. We have developed a method to accurately detect the velocity of object using a camera, based on a large-scale dataset collected over 20 years by the automotive manufacturer, SUBARU. The proposed method consists of three methods: an High Dynamic Range (HDR) detection method that fuses multiple stereo disparity images, a fusion method that combines the results of monocular…
UC San Diego（加州大学圣地亚哥分校）；ATG Robotics（ATG机器人公司）；University of Bristol（布里斯托尔大学）；Aristotle University of Thessaloniki（塞萨洛尼基亚里士多德大学）
As an important component of autonomous systems, autonomous car perception has had a big leap with recent advances in parallel computing architectures. With the use of tiny but full-feature embedded supercomputers, computer stereo vision has been prevalently applied in autonomous cars for depth perception. The two key aspects of computer stereo vision are speed and accuracy. They are both desirable but conflicting properties, as the algorithms with better disparity accuracy usually have higher computational complexity. Therefore, the main aim of developing a computer stereo vision algorithm fo…
Faculty of Electrical Engineering, University of Ljubljana（卢布尔雅那大学电气工程学院）
We address the problem of optical decalibration in mobile stereo camera setups, especially in context of autonomous vehicles. In real world conditions, an optical system is subject to various sources of anticipated and unanticipated mechanical stress (vibration, rough handling, collisions). Mechanical stress changes the geometry between the cameras that make up the stereo pair, and as a consequence, the pre-calculated epipolar geometry is no longer valid. Our method is based on optimization of camera geometry parameters and plugs directly into the output of the stereo matching algorithm. There…
Hong Kong University of Science and Technology（香港科技大学）
We propose a stereo vision-based approach for tracking the camera ego-motion and 3D semantic objects in dynamic autonomous driving scenarios. Instead of directly regressing the 3D bounding box using end-to-end approaches, we propose to use the easy-to-labeled 2D detection and discrete viewpoint classification together with a light-weight semantic inference method to obtain rough 3D object measurements. Based on the object-aware-aided camera pose tracking which is robust in dynamic environments, in combination with our novel dynamic object bundle adjustment (BA) approach to fuse temporal sparse…
Zhejiang University（浙江大学）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）
Pseudo-LiDAR 3D detectors have made remarkable progress in monocular 3D detection by enhancing the capability of perceiving depth with depth estimation networks, and using LiDAR-based 3D detection architectures. The advanced stereo 3D detectors can also accurately localize 3D objects. The gap in image-to-image generation for stereo views is much smaller than that in image-to-LiDAR generation. Motivated by this, we propose a Pseudo-Stereo 3D detection framework with three novel virtual view generation methods, including image-level generation, feature-level generation, and feature-clone, for de…
The Hong Kong University of Science and Technology（香港科技大学）
We propose a 3D object detection system with multi-sensor refinement in the context of autonomous driving. In our framework, the monocular camera serves as the fundamental sensor for 2D object proposal and initial 3D bounding box prediction. While the stereo cameras and LiDAR are treated as adaptive plug-in sensors to refine the 3D box localization performance. For each observed element in the raw measurement domain (e.g., pixels for stereo, 3D points for LiDAR), we model the local geometry as an instance vector representation, which indicates the 3D coordinate of each element respecting to th…
University of Chinese Academy of Sciences（中国科学院大学）
Although the recent image-based 3D object detection methods using Pseudo-LiDAR representation have shown great capabilities, a notable gap in efficiency and accuracy still exist compared with LiDAR-based methods. Besides, over-reliance on the stand-alone depth estimator, requiring a large number of pixel-wise annotations in the training stage and more computation in the inferencing stage, limits the scaling application in the real world. In this paper, we propose an efficient and accurate 3D object detection method from stereo images, named RTS3D. Different from the 3D occupancy space in the P…
The Hong Kong University of Science and Technology（香港科技大学）；DJI（大疆创新）
We propose a 3D object detection method for autonomous driving by fully exploiting the sparse and dense, semantic and geometry information in stereo imagery. Our method, called Stereo R-CNN, extends Faster R-CNN for stereo inputs to simultaneously detect and associate object in left and right images. We add extra branches after stereo Region Proposal Network (RPN) to predict sparse keypoints, viewpoints, and object dimensions, which are combined with 2D left-right boxes to calculate a coarse 3D object bounding box. We then recover the accurate 3D bounding box by a region-based photometric alig…
University of Michigan（密歇根大学）；TuSimple, Inc.（图森未来公司）
Dense stereo matching with deep neural networks is of great interest to the research community. Existing stereo matching networks typically use slow and computationally expensive 3D convolutions to improve the performance, which is not friendly to real-world applications such as autonomous driving. In this paper, we propose the Efficient Stereo Network (ESNet), which achieves high performance and efficient inference at the same time. ESNet relies only on 2D convolution and computes multi-scale cost volume efficiently using a warping-based method to improve the performance in regions with fine-…
Cornell University（康奈尔大学）；The Ohio State University（俄亥俄州立大学）
Detecting objects such as cars and pedestrians in 3D plays an indispensable role in autonomous driving. Existing approaches largely rely on expensive LiDAR sensors for accurate depth information. While recently pseudo-LiDAR has been introduced as a promising alternative, at a much lower cost based solely on stereo images, there is still a notable performance gap. In this paper we provide substantial advances to the pseudo-LiDAR framework through improvements in stereo depth estimation. Concretely, we adapt the stereo network architecture and loss function to be more aligned with accurate depth…
Sun Yat-sen University（中山大学）；University of Waterloo（滑铁卢大学）
Accurate and fast 3D object detection from point clouds is a key task in autonomous driving. Existing one-stage 3D object detection methods can achieve real-time performance, however, they are dominated by anchor-based detectors which are inefficient and require additional post-processing. In this paper, we eliminate anchors and model an object as a single point--the center point of its bounding box. Based on the center point, we propose an anchor-free CenterNet3D network that performs 3D object detection without anchors. Our CenterNet3D uses keypoint estimation to find center points and direc…
Institute for Information Transmission Problems (Kharkevich Institute) –IITP RAS（俄罗斯科学院信息传输问题研究所（哈尔凯维奇研究所））
In this work we present a method for performance evaluation of stereo vision based obstacle detection techniques that takes into account the specifics of road situation analysis to minimize the effort required to prepare a test dataset. This approach has been designed to be implemented in systems such as self-driving cars or driver assistance and can also be used as problem-oriented quality criterion for evaluation of stereo vision algorithms.
Dalian University of Technology（大连理工大学）；Beijing Institute of Technology（北京理工大学）；The University of Sydney（悉尼大学）
Being a crucial task of autonomous driving, Stereo matching has made great progress in recent years. Existing stereo matching methods estimate disparity instead of depth. They treat the disparity errors as the evaluation metric of the depth estimation errors, since the depth can be calculated from the disparity according to the triangulation principle. However, we find that the error of the depth depends not only on the error of the disparity but also on the depth range of the points. Therefore, even if the disparity error is low, the depth error is still large, especially for the distant poin…
University of Delaware（特拉华大学）；Vision Systems Inc.（视觉系统公司）
Digital Surface Model generation from satellite imagery is a difficult task that has been largely overlooked by the deep learning community. Stereo reconstruction techniques developed for terrestrial systems including self driving cars do not translate well to satellite imagery where image pairs vary considerably. In this work we present neural network tailored for Digital Surface Model generation, a ground truthing and training scheme which maximizes available hardware, and we present a comparison to existing methods. The resulting models are smooth, preserve boundaries, and enable further pr…
Peking University（北京大学）；National Innovation Institute of Defense Technology（国防科技创新研究院）
3D reconstruction has lately attracted increasing attention due to its wide application in many areas, such as autonomous driving, robotics and virtual reality. As a dominant technique in artificial intelligence, deep learning has been successfully adopted to solve various computer vision problems. However, deep learning for 3D reconstruction is still at its infancy due to its unique challenges and varying pipelines. To stimulate future research, this paper presents a review of recent progress in deep learning methods for Multi-view Stereo (MVS), which is considered as a crucial task of image-…
Disparity prediction from stereo images is essential to computer vision applications including autonomous driving, 3D model reconstruction, and object detection. To predict accurate disparity map, we propose a novel deep learning architecture for detectingthe disparity map from a rectified pair of stereo images, called MSDC-Net. Our MSDC-Net contains two modules: multi-scale fusion 2D convolution and multi-scale residual 3D convolution modules. The multi-scale fusion 2D convolution module exploits the potential multi-scale features, which extracts and fuses the different scale features by Dens…
EPFL（洛桑联邦理工学院）；Visual Intelligence for Transportation (VITA) lab（交通视觉智能实验室）
Monocular and stereo visions are cost-effective solutions for 3D human localization in the context of self-driving cars or social robots. However, they are usually developed independently and have their respective strengths and limitations. We propose a novel unified learning framework that leverages the strengths of both monocular and stereo cues for 3D human localization. Our method jointly (i) associates humans in left-right images, (ii) deals with occluded and distant cases in stereo settings by relying on the robustness of monocular cues, and (iii) tackles the intrinsic ambiguity of monoc…
University of Oxford（牛津大学）；University of Bologna（博洛尼亚大学）；FiveAI；Australian National University（澳大利亚国立大学）
Real world applications of stereo depth estimation require models that are robust to dynamic variations in the environment. Even though deep learning based stereo methods are successful, they often fail to generalize to unseen variations in the environment, making them less suitable for practical applications such as autonomous driving. In this work, we introduce a "learning-to-adapt" framework that enables deep stereo methods to continuously adapt to new target domains in an unsupervised manner. Specifically, our approach incorporates the adaptation procedure into the learning objective to ob…
Beihang University（北京航空航天大学）；National University of Singapore（新加坡国立大学）；Institute for Infocomm Research（资讯通信研究院）
Deep stereo matching has made significant progress in recent years. However, state-of-the-art methods are based on expensive 4D cost volume, which limits their use in real-world applications. To address this issue, 3D correlation maps and iterative disparity updates have been proposed. Regarding that in real-world platforms, such as self-driving cars and robots, the Lidar is usually installed. Thus we further introduce the sparse Lidar point into the iterative updates, which alleviates the burden of network updating the disparity from zero states. Furthermore, we propose training the network i…
Stereo matching plays an indispensable part in autonomous driving, robotics and 3D scene reconstruction. We propose a novel deep learning architecture, which called CFP-Net, a Cross-Form Pyramid stereo matching network for regressing disparity from a rectified pair of stereo images. The network consists of three modules: Multi-Scale 2D local feature extraction module, Cross-form spatial pyramid module and Multi-Scale 3D Feature Matching and Fusion module. The Multi-Scale 2D local feature extraction module can extract enough multi-scale features. The Cross-form spatial pyramid module aggregates…
Detecting small obstacles on the road ahead is a critical part of the driving task which has to be mastered by fully autonomous cars. In this paper, we present a method based on stereo vision to reliably detect such obstacles from a moving vehicle. The proposed algorithm performs statistical hypothesis tests in disparity space directly on stereo image data, assessing freespace and obstacle hypotheses on independent local patches. This detection approach does not depend on a global road model and handles both static and moving obstacles. For evaluation, we employ a novel lost-cargo image sequen…
Stereo matching is a key component of autonomous driving perception. Recent unsupervised stereo matching approaches have received adequate attention due to their advantage of not requiring disparity ground truth. These approaches, however, perform poorly near occlusions. To overcome this drawback, in this paper, we propose CoT-Stereo, a novel unsupervised stereo matching approach. Specifically, we adopt a co-teaching framework where two networks interactively teach each other about the occlusions in an unsupervised fashion, which greatly improves the robustness of unsupervised stereo matching.…
Visual perception plays an important role in autonomous driving. One of the primary tasks is object detection and identification. Since the vision sensor is rich in color and texture information, it can quickly and accurately identify various road information. The commonly used technique is based on extracting and calculating various features of the image. The recent development of deep learning-based method has better reliability and processing speed and has a greater advantage in recognizing complex elements. For depth estimation, vision sensor is also used for ranging due to their small siz…
3D object detection is an essential task in autonomous driving. Recent techniques excel with highly accurate detection rates, provided the 3D input data is obtained from precise but expensive LiDAR technology. Approaches based on cheaper monocular or stereo imagery data have, until now, resulted in drastically lower accuracies --- a gap that is commonly attributed to poor image-based depth estimation. However, in this paper we argue that it is not the quality of the data but its representation that accounts for the majority of the difference. Taking the inner workings of convolutional neural n…
Technion - Israel Institute of Technology（以色列理工学院）
Depth estimation is a cornerstone of a vast number of applications requiring 3D assessment of the environment, such as robotics, augmented reality, and autonomous driving to name a few. One prominent technique for depth estimation is stereo matching which has several advantages: it is considered more accessible than other depth-sensing technologies, can produce dense depth estimates in real-time, and has benefited greatly from the advances of deep learning in recent years. However, current techniques for depth estimation from stereoscopic images still suffer from a built-in drawback. To recons…
This paper describes a multimodal vision sensor that integrates three types of cameras, including a stereo camera, a polarization camera and a panoramic camera. Each sensor provides a specific dimension of information: the stereo camera measures depth per pixel, the polarization obtains the degree of polarization, and the panoramic camera captures a 360-degree landscape. Data fusion and advanced environment perception could be built upon the combination of sensors. Designed especially for autonomous driving, this vision sensor is shipped with a robust semantic segmentation network. In addition…
Tsinghua University（清华大学）；University of Toronto（多伦多大学）
The goal of this paper is to perform 3D object detection in the context of autonomous driving. Our method first aims at generating a set of high-quality 3D object proposals by exploiting stereo imagery. We formulate the problem as minimizing an energy function that encodes object size priors, placement of objects on the ground plane as well as several depth informed features that reason about free space, point cloud densities and distance to the ground. We then exploit a CNN on top of these proposals to perform object detection. In particular, we employ a convolutional neural net (CNN) that ex…
Australian Centre for Robotic Vision（澳大利亚机器人视觉中心）；Australian National University（澳大利亚国立大学）；Queensland University of Technology（昆士兰科技大学）
Current self-driving car systems operate well in sunny weather but struggle in adverse conditions. One of the most commonly encountered adverse conditions involves water on the road caused by rain, sleet, melting snow or flooding. While some advances have been made in using conventional RGB camera and LIDAR technology for detecting water hazards, other sources of information such as polarization offer a promising and potentially superior approach to this problem in terms of performance and cost. In this paper, we present a novel stereo-polarization system for detecting and tracking water hazar…
Motif Channel Opened in a White-Box: Stereo Matching via Motif Correlation Graph
School of Information Engineering, Guizhou University of Commerce（贵州商学院信息工程学院）；School of Software, Northwestern Polytechnical University（西北工业大学软件学院）；School of Computer Science and Engineering, South China University of Technology（华南理工大学计算机科学与工程学院）
针对基于学习的立体匹配中几何结构丢失与黑盒可解释性不足问题，提出 MoCha-V2，利用 MCG 和小波逆变换恢复结构，并在 Middlebury 取得第一。
DepthVanish: Optimizing Adversarial Interval Structures for Stereo-Depth-Invisible Patches
University of Alberta（阿尔伯塔大学）；VCIP, CS, Nankai University（VCIP、计算机科学系、南开大学）；NKIARI, Shenzhen Futian（NKIARI、深圳福田）；The University of Tokyo（东京大学）；CFAR and IHPC, Agency for Science, Technology and Research (A*STAR)（CFAR和IHPC、科技研究局（A*STAR））；Nanyang Technological University（南洋理工大学）
Accurate layout estimation is crucial for planning and navigation in robotics applications, such as self-driving. In this paper, we introduce the Stereo Bird's Eye ViewNetwork (SBEVNet), a novel supervised end-to-end framework for estimation of bird's eye view layout from a pair of stereo images. Although our network reuses some of the building blocks from the state-of-the-art deep learning networks for disparity estimation, we show that explicit depth estimation is neither sufficient nor necessary. Instead, the learning of a good internal bird's eye view feature representation is effective fo…
3D Semantic Scene Completion (SSC) can provide dense geometric and semantic scene representations, which can be applied in the field of autonomous driving and robotic systems. It is challenging to estimate the complete geometry and semantics of a scene solely from visual images, and accurate depth information is crucial for restoring 3D geometry. In this paper, we propose the first stereo SSC method named OccDepth, which fully exploits implicit depth information from stereo images (or RGBD images) to help the recovery of 3D geometric structures. The Stereo Soft Feature Assignment (Stereo-SFA) …
The Chinese University of Hong Kong（香港中文大学）；SenseTime Research（商汤科技研究院）
Stereo matching estimates the disparity between a rectified image pair, which is of great importance to depth sensing, autonomous driving, and other related tasks. Previous works built cost volumes with cross-correlation or concatenation of left and right features across all disparity levels, and then a 2D or 3D convolutional neural network is utilized to regress the disparity maps. In this paper, we propose to construct the cost volume by group-wise correlation. The left features and the right features are divided into groups along the channel dimension, and correlation maps are computed amon…
Nankai University（南开大学）；RIKEN（理化学研究所）；Skolkovo Institute of Science and Technology（斯科尔科沃科学技术学院）；Hangzhou Dianzi University（杭州电子科技大学）；Nicolaus Copernicus University（哥白尼大学）；NVIDIA Corporation（英伟达公司）
A large number of autonomous driving tasks need high-definition stereo images, which requires a large amount of storage space. Efficiently executing lossless compression has become a practical problem. Commonly, it is hard to make accurate probability estimates for each pixel. To tackle this, we propose L3C-Stereo, a multi-scale lossless compression model consisting of two main modules: the warping module and the probability estimation module. The warping module takes advantage of two view feature maps from the same domain to generate a disparity map, which is used to reconstruct the right vie…
We explore the problem of real-time stereo matching on high-res imagery. Many state-of-the-art (SOTA) methods struggle to process high-res imagery because of memory constraints or speed limitations. To address this issue, we propose an end-to-end framework that searches for correspondences incrementally over a coarse-to-fine hierarchy. Because high-res stereo datasets are relatively rare, we introduce a dataset with high-res stereo pairs for both training and evaluation. Our approach achieved SOTA performance on Middlebury-v3 and KITTI-15 while running significantly faster than its competitors…
Dense 3D maps from wide-angle cameras is beneficial to robotics applications such as navigation and autonomous driving. In this work, we propose a real-time dense 3D mapping method for fisheye cameras without explicit rectification and undistortion. We extend the conventional variational stereo method by constraining the correspondence search along the epipolar curve using a trajectory field induced by camera motion. We also propose a fast way of generating the trajectory field without increasing the processing time compared to conventional rectified methods. With our implementation, we were a…
Existing vision systems for autonomous driving or robots are sensitive to waterdrops adhered to windows or camera lenses. Most recent waterdrop removal approaches take a single image as input and often fail to recover the missing content behind waterdrops faithfully. Thus, we propose a learning-based model for waterdrop removal with stereo images. To better detect and remove waterdrops from stereo images, we propose a novel row-wise dilated attention module to enlarge attention's receptive field for effective information propagation between the two stereo images. In addition, we propose an att…
Australia National University（澳大利亚国立大学）；NVIDIA（英伟达）；Northwestern Polytechnical University（西北工业大学）
Although deep learning-based methods have dominated stereo matching leaderboards by yielding unprecedented disparity accuracy, their inference time is typically slow, on the order of seconds for a pair of 540p images. The main reason is that the leading methods employ time-consuming 3D convolutions applied to a 4D feature volume. A common way to speed up the computation is to downsample the feature volume, but this loses high-frequency details. To overcome these challenges, we propose a \emph{displacement-invariant cost computation module} to compute the matching costs without needing a 4D fea…
