Monocular 3D scene understanding tasks, such as object size estimation, heading angle estimation and 3D localization, is challenging. Successful modern day methods for 3D scene understanding require the use of a 3D sensor. On the other hand, single image based methods have significantly worse performance. In this work, we aim at bridging the performance gap between 3D sensing and 2D sensing for 3D object detection by enhancing LiDAR-based algorithms to work with single image input. Specifically, we perform monocular depth estimation and lift the input image to a point cloud representation, which we call pseudo-LiDAR point cloud. Then we can train a LiDAR-based 3D detection network with our pseudo-LiDAR end-to-end. Following the pipeline of two-stage 3D detection algorithms, we detect 2D object proposals in the input image and extract a point cloud frustum from the pseudo-LiDAR for each proposal. Then an oriented 3D bounding box is detected for each frustum. To handle the large amount of noise in the pseudo-LiDAR, we propose two innovations: (1) use a 2D-3D bounding box consistency constraint, adjusting the predicted 3D bounding box to have a high overlap with its corresponding 2D proposal after projecting onto the image; (2) use the instance mask instead of the bounding box as the representation of 2D proposals, in order to reduce the number of points not belonging to the object in the point cloud frustum. Through our evaluation on the KITTI benchmark, we achieve the top-ranked performance on both bird's eye view and 3D object detection among all monocular methods, effectively quadrupling the performance over previous state-of-the-art. Our code is available at https://github.com/xinshuoweng/Mono3D_PLiDAR.
Supervised learning methods to infer (hypothesize) depth of a scene from a single image require costly per-pixel ground-truth. We follow a geometric approach that exploits abundant stereo imagery to learn a model to hypothesize scene structure without direct supervision. Although we train a network with stereo pairs, we only require a single image at test time to hypothesize disparity or depth. We propose a novel objective function that exploits the bilateral cyclic relationship between the left and right disparities and we introduce an adaptive regularization scheme that allows the network to handle both the co-visible and occluded regions in a stereo pair. This process ultimately produces a model to generate hypotheses for the 3-dimensional structure of the scene as viewed in a single image. When used to generate a single (most probable) estimate of depth, our method outperforms state-of-the-art unsupervised monocular depth prediction methods on the KITTI benchmarks. We show that our method generalizes well by applying our models trained on KITTI to the Make3d dataset.
In this work, we consider transferring the structure information from large networks to compact ones for dense prediction tasks in computer vision. Previous knowledge distillation strategies used for dense prediction tasks often directly borrow the distillation scheme for image classification and perform knowledge distillation for each pixel separately, leading to sub-optimal performance. Here we propose to distill structured knowledge from large networks to compact networks, taking into account the fact that dense prediction is a structured prediction problem. Specifically, we study two structured distillation schemes: i) pair-wise distillation that distills the pair-wise similarities by building a static graph; and ii) holistic distillation that uses adversarial training to distill holistic knowledge. The effectiveness of our knowledge distillation approaches is demonstrated by experiments on three dense prediction tasks: semantic segmentation, depth estimation and object detection. Code is available at: https://git.io/StructKD
While learning based depth estimation from images/videos has achieved substantial progress, there still exist intrinsic limitations. Supervised methods are limited by a small amount of ground truth or labeled data and unsupervised methods for monocular videos are mostly based on the static scene assumption, not performing well on real world scenarios with the presence of dynamic objects. In this paper, we propose a new learning based method consisting of DepthNet, PoseNet and Region Deformer Networks (RDN) to estimate depth from unconstrained monocular videos without ground truth supervision. The core contribution lies in RDN for proper handling of rigid and non-rigid motions of various objects such as rigidly moving cars and deformable humans. In particular, a deformation based motion representation is proposed to model individual object motion on 2D images. This representation enables our method to be applicable to diverse unconstrained monocular videos. Our method can not only achieve the state-of-the-art results on standard benchmarks KITTI and Cityscapes, but also show promising results on a crowded pedestrian tracking dataset, which demonstrates the effectiveness of the deformation based motion representation. Code and trained models are available at https://github.com/haofeixu/rdn4depth.
We present a self-supervised approach to training convolutional neural networks for dense depth estimation from monocular endoscopy data without a priori modeling of anatomy or shading. Our method only requires monocular endoscopic videos and a multi-view stereo method, e.g., structure from motion, to supervise learning in a sparse manner. Consequently, our method requires neither manual labeling nor patient computed tomography (CT) scan in the training and application phases. In a cross-patient experiment using CT scans as groundtruth, the proposed method achieved submillimeter mean residual error. In a comparison study to recent self-supervised depth estimation methods designed for natural video on in vivo sinus endoscopy data, we demonstrate that the proposed approach outperforms the previous methods by a large margin. The source code for this work is publicly available online at https://github.com/lppllppl920/EndoscopyDepthEstimation-Pytorch.
This work proposes a new method to accurately complete sparse LiDAR maps guided by RGB images. For autonomous vehicles and robotics the use of LiDAR is indispensable in order to achieve precise depth predictions. A multitude of applications depend on the awareness of their surroundings, and use depth cues to reason and react accordingly. On the one hand, monocular depth prediction methods fail to generate absolute and precise depth maps. On the other hand, stereoscopic approaches are still significantly outperformed by LiDAR based approaches. The goal of the depth completion task is to generate dense depth predictions from sparse and irregular point clouds which are mapped to a 2D plane. We propose a new framework which extracts both global and local information in order to produce proper depth maps. We argue that simple depth completion does not require a deep network. However, we additionally propose a fusion method with RGB guidance from a monocular camera in order to leverage object information and to correct mistakes in the sparse input. This improves the accuracy significantly. Moreover, confidence masks are exploited in order to take into account the uncertainty in the depth predictions from each modality. This fusion method outperforms the state-of-the-art and ranks first on the KITTI depth completion benchmark. Our code with visualizations is available.
This paper presents a novel method, MaskMVS, to solve depth estimation for unstructured multi-view image-pose pairs. In the plane-sweep procedure, the depth planes are sampled by histogram matching that ensures covering the depth range of interest. Unlike other plane-sweep methods, we do not rely on a cost metric to explicitly build the cost volume, but instead infer a multiplane mask representation which regularizes the learning. Compared to many previous approaches, we show that our method is lightweight and generalizes well without requiring excessive training. We outperform the current state-of-the-art and show results on the sun3d, scenes11, MVS, and RGBD test data sets.
Depth estimation is a traditional computer vision task, which plays a crucial role in understanding 3D scene geometry. Recently, deep-convolutional-neural-networks based methods have achieved promising results in the monocular depth estimation field. Specifically, the framework that combines the multi-scale features extracted by the dilated convolution based block (atrous spatial pyramid pooling, ASPP) has gained the significant improvement in the dense labeling task. However, the discretized and predefined dilation rates cannot capture the continuous context information that differs in diverse scenes and easily introduce the grid artifacts in depth estimation. In this paper, we propose an attention-based context aggregation network (ACAN) to tackle these difficulties. Based on the self-attention model, ACAN adaptively learns the task-specific similarities between pixels to model the context information. First, we recast the monocular depth estimation as a dense labeling multi-class classification problem. Then we propose a soft ordinal inference to transform the predicted probabilities to continuous depth values, which can reduce the discretization error (about 1% decrease in RMSE). Second, the proposed ACAN aggregates both the image-level and pixel-level context information for depth estimation, where the former expresses the statistical characteristic of the whole image and the latter extracts the long-range spatial dependencies for each pixel. Third, for further reducing the inconsistency between the RGB image and depth map, we construct an attention loss to minimize their information entropy. We evaluate on public monocular depth-estimation benchmark datasets (including NYU Depth V2, KITTI). The experiments demonstrate the superiority of our proposed ACAN and achieve the competitive results with the state of the arts.
We introduce a convolutional neural network model for unsupervised learning of depth and ego-motion from cylindrical panoramic video. Panoramic depth estimation is an important technology for applications such as virtual reality, 3D modeling, and autonomous robotic navigation. In contrast to previous approaches for applying convolutional neural networks to panoramic imagery, we use the cylindrical panoramic projection which allows for the use of the traditional CNN layers such as convolutional filters and max pooling without modification. Our evaluation of synthetic and real data shows that unsupervised learning of depth and ego-motion on cylindrical panoramic images can produce high-quality depth maps and that an increased field-of-view improves ego-motion estimation accuracy. We also introduce Headcam, a novel dataset of panoramic video collected from a helmet-mounted camera while biking in an urban setting.
Accurate depth estimation from images is a fundamental task in many applications including scene understanding and reconstruction. Existing solutions for depth estimation often produce blurry approximations of low resolution. This paper presents a convolutional neural network for computing a high-resolution depth map given a single RGB image with the help of transfer learning. Following a standard encoder-decoder architecture, we leverage features extracted using high performing pre-trained networks when initializing our encoder along with augmentation and training strategies that lead to more accurate results. We show how, even for a very simple decoder, our method is able to achieve detailed high-resolution depth maps. Our network, with fewer parameters and training iterations, outperforms state-of-the-art on two datasets and also produces qualitatively better results that capture object boundaries more faithfully. Code and corresponding pre-trained weights are made publicly available.
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 networks into consideration, we propose to convert image-based depth maps to pseudo-LiDAR representations --- essentially mimicking the LiDAR signal. With this representation we can apply different existing LiDAR-based detection algorithms. On the popular KITTI benchmark, our approach achieves impressive improvements over the existing state-of-the-art in image-based performance --- raising the detection accuracy of objects within the 30m range from the previous state-of-the-art of 22% to an unprecedented 74%. At the time of submission our algorithm holds the highest entry on the KITTI 3D object detection leaderboard for stereo-image-based approaches. Our code is publicly available at https://github.com/mileyan/pseudo_lidar.
Unsupervised learning for geometric perception (depth, optical flow, etc.) is of great interest to autonomous systems. Recent works on unsupervised learning have made considerable progress on perceiving geometry; however, they usually ignore the coherence of objects and perform poorly under scenarios with dark and noisy environments. In contrast, supervised learning algorithms, which are robust, require large labeled geometric dataset. This paper introduces SIGNet, a novel framework that provides robust geometry perception without requiring geometrically informative labels. Specifically, SIGNet integrates semantic information to make depth and flow predictions consistent with objects and robust to low lighting conditions. SIGNet is shown to improve upon the state-of-the-art unsupervised learning for depth prediction by 30% (in squared relative error). In particular, SIGNet improves the dynamic object class performance by 39% in depth prediction and 29% in flow prediction. Our code will be made available at https://github.com/mengyuest/SIGNet
We propose DeepV2D, an end-to-end deep learning architecture for predicting depth from video. DeepV2D combines the representation ability of neural networks with the geometric principles governing image formation. We compose a collection of classical geometric algorithms, which are converted into trainable modules and combined into an end-to-end differentiable architecture. DeepV2D interleaves two stages: motion estimation and depth estimation. During inference, motion and depth estimation are alternated and converge to accurate depth. Code is available https://github.com/princeton-vl/DeepV2D.
Dot-product attention has wide applications in computer vision and natural language processing. However, its memory and computational costs grow quadratically with the input size. Such growth prohibits its application on high-resolution inputs. To remedy this drawback, this paper proposes a novel efficient attention mechanism equivalent to dot-product attention but with substantially less memory and computational costs. Its resource efficiency allows more widespread and flexible integration of attention modules into a network, which leads to better accuracies. Empirical evaluations demonstrated the effectiveness of its advantages. Efficient attention modules brought significant performance boosts to object detectors and instance segmenters on MS-COCO 2017. Further, the resource efficiency democratizes attention to complex models, where high costs prohibit the use of dot-product attention. As an exemplar, a model with efficient attention achieved state-of-the-art accuracies for stereo depth estimation on the Scene Flow dataset. Code is available at https://github.com/cmsflash/efficient-attention.
Detecting and localizing objects in the real 3D space, which plays a crucial role in scene understanding, is particularly challenging given only a single RGB image due to the geometric information loss during imagery projection. We propose MonoGRNet for the amodal 3D object detection from a monocular RGB image via geometric reasoning in both the observed 2D projection and the unobserved depth dimension. MonoGRNet is a single, unified network composed of four task-specific subnetworks, responsible for 2D object detection, instance depth estimation (IDE), 3D localization and local corner regression. Unlike the pixel-level depth estimation that needs per-pixel annotations, we propose a novel IDE method that directly predicts the depth of the targeting 3D bounding box's center using sparse supervision. The 3D localization is further achieved by estimating the position in the horizontal and vertical dimensions. Finally, MonoGRNet is jointly learned by optimizing the locations and poses of the 3D bounding boxes in the global context. We demonstrate that MonoGRNet achieves state-of-the-art performance on challenging datasets.
In this paper, we study 1-bit convolutional neural networks (CNNs), of which both the weights and activations are binary. While efficient, the lacking of representational capability and the training difficulty impede 1-bit CNNs from performing as well as real-valued networks. We propose Bi-Real net with a novel training algorithm to tackle these two challenges. To enhance the representational capability, we propagate the real-valued activations generated by each 1-bit convolution via a parameter-free shortcut. To address the training difficulty, we propose a training algorithm using a tighter approximation to the derivative of the sign function, a magnitude-aware gradient for weight updating, a better initialization method, and a two-step scheme for training a deep network. Experiments on ImageNet show that an 18-layer Bi-Real net with the proposed training algorithm achieves 56.4% top-1 classification accuracy, which is 10% higher than the state-of-the-arts (e.g., XNOR-Net) with greater memory saving and lower computational cost. Bi-Real net is also the first to scale up 1-bit CNNs to an ultra-deep network with 152 layers, and achieves 64.5% top-1 accuracy on ImageNet. A 50-layer Bi-Real net shows comparable performance to a real-valued network on the depth estimation task with only a 0.3% accuracy gap.
Many applications of stereo depth estimation in robotics require the generation of accurate disparity maps in real time under significant computational constraints. Current state-of-the-art algorithms force a choice between either generating accurate mappings at a slow pace, or quickly generating inaccurate ones, and additionally these methods typically require far too many parameters to be usable on power- or memory-constrained devices. Motivated by these shortcomings, we propose a novel approach for disparity prediction in the anytime setting. In contrast to prior work, our end-to-end learned approach can trade off computation and accuracy at inference time. Depth estimation is performed in stages, during which the model can be queried at any time to output its current best estimate. Our final model can process 1242$ \times $375 resolution images within a range of 10-35 FPS on an NVIDIA Jetson TX2 module with only marginal increases in error -- using two orders of magnitude fewer parameters than the most competitive baseline. The source code is available at https://github.com/mileyan/AnyNet .
Automated design of neural network architectures tailored for a specific task is an extremely promising, albeit inherently difficult, avenue to explore. While most results in this domain have been achieved on image classification and language modelling problems, here we concentrate on dense per-pixel tasks, in particular, semantic image segmentation using fully convolutional networks. In contrast to the aforementioned areas, the design choices of a fully convolutional network require several changes, ranging from the sort of operations that need to be used---e.g., dilated convolutions---to a solving of a more difficult optimisation problem. In this work, we are particularly interested in searching for high-performance compact segmentation architectures, able to run in real-time using limited resources. To achieve that, we intentionally over-parameterise the architecture during the training time via a set of auxiliary cells that provide an intermediate supervisory signal and can be omitted during the evaluation phase. The design of the auxiliary cell is emitted by a controller, a neural network with the fixed structure trained using reinforcement learning. More crucially, we demonstrate how to efficiently search for these architectures within limited time and computational budgets. In particular, we rely on a progressive strategy that terminates non-promising architectures from being further trained, and on Polyak averaging coupled with knowledge distillation to speed-up the convergence. Quantitatively, in 8 GPU-days our approach discovers a set of architectures performing on-par with state-of-the-art among compact models on the semantic segmentation, pose estimation and depth prediction tasks. Code will be made available here: https://github.com/drsleep/nas-segm-pytorch
Data-driven algorithms have surpassed traditional techniques in almost every aspect in robotic vision problems. Such algorithms need vast amounts of quality data to be able to work properly after their training process. Gathering and annotating that sheer amount of data in the real world is a time-consuming and error-prone task. Those problems limit scale and quality. Synthetic data generation has become increasingly popular since it is faster to generate and automatic to annotate. However, most of the current datasets and environments lack realism, interactions, and details from the real world. UnrealROX is an environment built over Unreal Engine 4 which aims to reduce that reality gap by leveraging hyperrealistic indoor scenes that are explored by robot agents which also interact with objects in a visually realistic manner in that simulated world. Photorealistic scenes and robots are rendered by Unreal Engine into a virtual reality headset which captures gaze so that a human operator can move the robot and use controllers for the robotic hands; scene information is dumped on a per-frame basis so that it can be reproduced offline to generate raw data and ground truth annotations. This virtual reality environment enables robotic vision researchers to generate realistic and visually plausible data with full ground truth for a wide variety of problems such as class and instance semantic segmentation, object detection, depth estimation, visual grasping, and navigation.
Learning to estimate 3D geometry in a single frame and optical flow from consecutive frames by watching unlabeled videos via deep convolutional network has made significant progress recently. Current state-of-the-art (SoTA) methods treat the two tasks independently. One typical assumption of the existing depth estimation methods is that the scenes contain no independent moving objects. while object moving could be easily modeled using optical flow. In this paper, we propose to address the two tasks as a whole, i.e. to jointly understand per-pixel 3D geometry and motion. This eliminates the need of static scene assumption and enforces the inherent geometrical consistency during the learning process, yielding significantly improved results for both tasks. We call our method as "Every Pixel Counts++" or "EPC++". Specifically, during training, given two consecutive frames from a video, we adopt three parallel networks to predict the camera motion (MotionNet), dense depth map (DepthNet), and per-pixel optical flow between two frames (OptFlowNet) respectively. The three types of information are fed into a holistic 3D motion parser (HMP), and per-pixel 3D motion of both rigid background and moving objects are disentangled and recovered. Comprehensive experiments were conducted on datasets with different scenes, including driving scenario (KITTI 2012 and KITTI 2015 datasets), mixed outdoor/indoor scenes (Make3D) and synthetic animation (MPI Sintel dataset). Performance on the five tasks of depth estimation, optical flow estimation, odometry, moving object segmentation and scene flow estimation shows that our approach outperforms other SoTA methods. Code will be available at: https://github.com/chenxuluo/EPC.