Single-view depth estimation from omnidirectional images has gained popularity with its wide range of applications such as autonomous driving and scene reconstruction. Although data-driven learning-based methods demonstrate significant potential in this field, scarce training data and ineffective 360 estimation algorithms are still two key limitations hindering accurate estimation across diverse domains. In this work, we first establish a large-scale dataset with varied settings called Depth360 to tackle the training data problem. This is achieved by exploring the use of a plenteous source of data, 360 videos from the internet, using a test-time training method that leverages unique information in each omnidirectional sequence. With novel geometric and temporal constraints, our method generates consistent and convincing depth samples to facilitate single-view estimation. We then propose an end-to-end two-branch multi-task learning network, SegFuse, that mimics the human eye to effectively learn from the dataset and estimate high-quality depth maps from diverse monocular RGB images. With a peripheral branch that uses equirectangular projection for depth estimation and a foveal branch that uses cubemap projection for semantic segmentation, our method predicts consistent global depth while maintaining sharp details at local regions. Experimental results show favorable performance against the state-of-the-art methods.
The ongoing biodiversity crisis calls for accurate estimation of animal density and abundance to identify sources of biodiversity decline and effectiveness of conservation interventions. Camera traps together with abundance estimation methods are often employed for this purpose. The necessary distances between camera and observed animals are traditionally derived in a laborious, fully manual or semi-automatic process. Both approaches require reference image material, which is both difficult to acquire and not available for existing datasets. We propose a fully automatic approach we call AUtomated DIstance esTimation (AUDIT) to estimate camera-to-animal distances. We leverage existing state-of-the-art relative monocular depth estimation and combine it with a novel alignment procedure to estimate metric distances. AUDIT is fully automated and requires neither the comparison of observations in camera trap imagery with reference images nor capturing of reference image material at all. AUDIT therefore relieves biologists and ecologists from a significant workload. We evaluate AUDIT on a zoo scenario dataset unseen during training where we achieve a mean absolute distance estimation error over all animal instances of only 0.9864 meters and mean relative error (REL) of 0.113. The code and usage instructions are available at https://github.com/PJ-cs/DistanceEstimationTracking
In this paper, HeadPosr is proposed to predict the head poses using a single RGB image. \textit{HeadPosr} uses a novel architecture which includes a transformer encoder. In concrete, it consists of: (1) backbone; (2) connector; (3) transformer encoder; (4) prediction head. The significance of using a transformer encoder for HPE is studied. An extensive ablation study is performed on varying the (1) number of encoders; (2) number of heads; (3) different position embeddings; (4) different activations; (5) input channel size, in a transformer used in HeadPosr. Further studies on using: (1) different backbones, (2) using different learning rates are also shown. The elaborated experiments and ablations studies are conducted using three different open-source widely used datasets for HPE, i.e., 300W-LP, AFLW2000, and BIWI datasets. Experiments illustrate that \textit{HeadPosr} outperforms all the state-of-art methods including both the landmark-free and the others based on using landmark or depth estimation on the AFLW2000 dataset and BIWI datasets when trained with 300W-LP. It also outperforms when averaging the results from the compared datasets, hence setting a benchmark for the problem of HPE, also demonstrating the effectiveness of using transformers over the state-of-the-art.
We benchmark the robustness of maximum likelihood based uncertainty estimation methods to outliers in training data for regression tasks. Outliers or noisy labels in training data results in degraded performances as well as incorrect estimation of uncertainty. We propose the use of a heavy-tailed distribution (Laplace distribution) to improve the robustness to outliers. This property is evaluated using standard regression benchmarks and on a high-dimensional regression task of monocular depth estimation, both containing outliers. In particular, heavy-tailed distribution based maximum likelihood provides better uncertainty estimates, better separation in uncertainty for out-of-distribution data, as well as better detection of adversarial attacks in the presence of outliers.
In this paper, we propose a learning-based method for predicting dense depth values of a scene from a monocular omnidirectional image. An omnidirectional image has a full field-of-view, providing much more complete descriptions of the scene than perspective images. However, fully-convolutional networks that most current solutions rely on fail to capture rich global contexts from the panorama. To address this issue and also the distortion of equirectangular projection in the panorama, we propose Cubemap Vision Transformers (CViT), a new transformer-based architecture that can model long-range dependencies and extract distortion-free global features from the panorama. We show that cubemap vision transformers have a global receptive field at every stage and can provide globally coherent predictions for spherical signals. To preserve important local features, we further design a convolution-based branch in our pipeline (dubbed GLPanoDepth) and fuse global features from cubemap vision transformers at multiple scales. This global-to-local strategy allows us to fully exploit useful global and local features in the panorama, achieving state-of-the-art performance in panoramic depth estimation.
Autonomous checkout systems rely on visual and sensory inputs to carry out fine-grained scene understanding in retail environments. Retail environments present unique challenges compared to typical indoor scenes owing to the vast number of densely packed, unique yet similar objects. The problem becomes even more difficult when only RGB input is available, especially for data-hungry tasks such as instance segmentation. To address the lack of datasets for retail, we present StandardSim, a large-scale photorealistic synthetic dataset featuring annotations for semantic segmentation, instance segmentation, depth estimation, and object detection. Our dataset provides multiple views per scene, enabling multi-view representation learning. Further, we introduce a novel task central to autonomous checkout called change detection, requiring pixel-level classification of takes, puts and shifts in objects over time. We benchmark widely-used models for segmentation and depth estimation on our dataset, show that our test set constitutes a difficult benchmark compared to current smaller-scale datasets and that our training set provides models with crucial information for autonomous checkout tasks.
The advent of autonomous driving and advanced driver assistance systems necessitates continuous developments in computer vision for 3D scene understanding. Self-supervised monocular depth estimation, a method for pixel-wise distance estimation of objects from a single camera without the use of ground truth labels, is an important task in 3D scene understanding. However, existing methods for this task are limited to convolutional neural network (CNN) architectures. In contrast with CNNs that use localized linear operations and lose feature resolution across the layers, vision transformers process at constant resolution with a global receptive field at every stage. While recent works have compared transformers against their CNN counterparts for tasks such as image classification, no study exists that investigates the impact of using transformers for self-supervised monocular depth estimation. Here, we first demonstrate how to adapt vision transformers for self-supervised monocular depth estimation. Thereafter, we compare the transformer and CNN-based architectures for their performance on KITTI depth prediction benchmarks, as well as their robustness to natural corruptions and adversarial attacks, including when the camera intrinsics are unknown. Our study demonstrates how transformer-based architecture, though lower in run-time efficiency, achieves comparable performance while being more robust and generalizable.
We propose an approach to semantic segmentation that achieves state-of-the-art supervised performance when applied in a zero-shot setting. It thus achieves results equivalent to those of the supervised methods, on each of the major semantic segmentation datasets, without training on those datasets. This is achieved by replacing each class label with a vector-valued embedding of a short paragraph that describes the class. The generality and simplicity of this approach enables merging multiple datasets from different domains, each with varying class labels and semantics. The resulting merged semantic segmentation dataset of over 2 Million images enables training a model that achieves performance equal to that of state-of-the-art supervised methods on 7 benchmark datasets, despite not using any images therefrom. By fine-tuning the model on standard semantic segmentation datasets, we also achieve a significant improvement over the state-of-the-art supervised segmentation on NYUD-V2 and PASCAL-context at 60% and 65% mIoU, respectively. Based on the closeness of language embeddings, our method can even segment unseen labels. Extensive experiments demonstrate strong generalization to unseen image domains and unseen labels, and that the method enables impressive performance improvements in downstream applications, including depth estimation and instance segmentation.
Existing monocular depth estimation methods have achieved excellent robustness in diverse scenes, but they can only retrieve affine-invariant depth, up to an unknown scale and shift. However, in some video-based scenarios such as video depth estimation and 3D scene reconstruction from a video, the unknown scale and shift residing in per-frame prediction may cause the depth inconsistency. To solve this problem, we propose a locally weighted linear regression method to recover the scale and shift with very sparse anchor points, which ensures the scale consistency along consecutive frames. Extensive experiments show that our method can boost the performance of existing state-of-the-art approaches by 50% at most over several zero-shot benchmarks. Besides, we merge over 6.3 million RGBD images to train strong and robust depth models. Our produced ResNet50-backbone model even outperforms the state-of-the-art DPT ViT-Large model. Combining with geometry-based reconstruction methods, we formulate a new dense 3D scene reconstruction pipeline, which benefits from both the scale consistency of sparse points and the robustness of monocular methods. By performing the simple per-frame prediction over a video, the accurate 3D scene shape can be recovered.
Omnidirectional 3D information is essential for a wide range of applications such as Virtual Reality, Autonomous Driving, Robotics, etc. In this paper, we propose a novel, model-agnostic, two-stage pipeline for omnidirectional monocular depth estimation. Our proposed framework PanoDepth takes one 360 image as input, produces one or more synthesized views in the first stage, and feeds the original image and the synthesized images into the subsequent stereo matching stage. In the second stage, we propose a differentiable Spherical Warping Layer to handle omnidirectional stereo geometry efficiently and effectively. By utilizing the explicit stereo-based geometric constraints in the stereo matching stage, PanoDepth can generate dense high-quality depth. We conducted extensive experiments and ablation studies to evaluate PanoDepth with both the full pipeline as well as the individual modules in each stage. Our results show that PanoDepth outperforms the state-of-the-art approaches by a large margin for 360 monocular depth estimation.
Hand-held light field (LF) cameras have unique advantages in computer vision such as 3D scene reconstruction and depth estimation. However, the related applications are limited by the ultra-small baseline, e.g., leading to the extremely low depth resolution in reconstruction. To solve this problem, we propose to rectify LF to obtain a large baseline. Specifically, the proposed method aligns two LFs captured by two hand-held LF cameras with a random relative pose, and extracts the corresponding row-aligned sub-aperture images (SAIs) to obtain an LF with a large baseline. For an accurate rectification, a method for pose estimation is also proposed, where the relative rotation and translation between the two LF cameras are estimated. The proposed pose estimation minimizes the degree of freedom (DoF) in the LF-point-LF-point correspondence model and explicitly solves this model in a linear way. The proposed pose estimation outperforms the state-of-the-art algorithms by providing more accurate results to support rectification. The significantly improved depth resolution in 3D reconstruction demonstrates the effectiveness of the proposed LF rectification.
This paper addresses the problem of mobile robot manipulation using object detection. Our approach uses detection and control as complimentary functions that learn from real-world interactions. We develop an end-to-end manipulation method based solely on detection and introduce Task-focused Few-shot Object Detection (TFOD) to learn new objects and settings. Our robot collects its own training data and automatically determines when to retrain detection to improve performance across various subtasks (e.g., grasping). Notably, detection training is low-cost, and our robot learns to manipulate new objects using as few as four clicks of annotation. In physical experiments, our robot learns visual control from a single click of annotation and a novel update formulation, manipulates new objects in clutter and other mobile settings, and achieves state-of-the-art results on an existing visual servo control and depth estimation benchmark. Finally, we develop a TFOD Benchmark to support future object detection research for robotics: https://github.com/griffbr/tfod.
Real-time estimation of actual object depth is an essential module for various autonomous system tasks such as 3D reconstruction, scene understanding and condition assessment. During the last decade of machine learning, extensive deployment of deep learning methods to computer vision tasks has yielded approaches that succeed in achieving realistic depth synthesis out of a simple RGB modality. Most of these models are based on paired RGB-depth data and/or the availability of video sequences and stereo images. The lack of sequences, stereo data and RGB-depth pairs makes depth estimation a fully unsupervised single-image transfer problem that has barely been explored so far. This study builds on recent advances in the field of generative neural networks in order to establish fully unsupervised single-shot depth estimation. Two generators for RGB-to-depth and depth-to-RGB transfer are implemented and simultaneously optimized using the Wasserstein-1 distance, a novel perceptual reconstruction term and hand-crafted image filters. We comprehensively evaluate the models using industrial surface depth data as well as the Texas 3D Face Recognition Database, the CelebAMask-HQ database of human portraits and the SURREAL dataset that records body depth. For each evaluation dataset the proposed method shows a significant increase in depth accuracy compared to state-of-the-art single-image transfer methods.
To support minimally-invasive intraoperative mitral valve repair, quantitative measurements from the valve can be obtained using an infra-red tracked stylus. It is desirable to view such manually measured points together with the endoscopic image for further assistance. Therefore, hand-eye calibration is required that links both coordinate systems and is a prerequisite to project the points onto the image plane. A complementary approach to this is to use a vision-based endoscopic stereo-setup to detect and triangulate points of interest, to obtain the 3D coordinates. In this paper, we aim to compare both approaches on a rigid phantom and two patient-individual silicone replica which resemble the intraoperative scenario. The preliminary results indicate that 3D landmark estimation, either labeled manually or through partly automated detection with a deep learning approach, provides more accurate triangulated depth measurements when performed with a tailored image-based method than with stylus measurements.
Reference-guided image inpainting restores image pixels by leveraging the content from another single reference image. The primary challenge is how to precisely place the pixels from the reference image into the hole region. Therefore, understanding the 3D geometry that relates pixels between two views is a crucial step towards building a better model. Given the complexity of handling various types of reference images, we focus on the scenario where the images are captured by freely moving the same camera around. Compared to the previous work, we propose a principled approach that does not make heuristic assumptions about the planarity of the scene. We leverage a monocular depth estimate and predict relative pose between cameras, then align the reference image to the target by a differentiable 3D reprojection and a joint optimization of relative pose and depth map scale and offset. Our approach achieves state-of-the-art performance on both RealEstate10K and MannequinChallenge dataset with large baselines, complex geometry and extreme camera motions. We experimentally verify our approach is also better at handling large holes.
In this paper, we introduce a deep multi-view stereo (MVS) system that jointly predicts depths, surface normals and per-view confidence maps. The key to our approach is a novel solver that iteratively solves for per-view depth map and normal map by optimizing an energy potential based on the locally planar assumption. Specifically, the algorithm updates depth map by propagating from neighboring pixels with slanted planes, and updates normal map with local probabilistic plane fitting. Both two steps are monitored by a customized confidence map. This solver is not only effective as a post-processing tool for plane-based depth refinement and completion, but also differentiable such that it can be efficiently integrated into deep learning pipelines. Our multi-view stereo system employs multiple optimization steps of the solver over the initial prediction of depths and surface normals. The whole system can be trained end-to-end, decoupling the challenging problem of matching pixels within poorly textured regions from the cost-volume based neural network. Experimental results on ScanNet and RGB-D Scenes V2 demonstrate state-of-the-art performance of the proposed deep MVS system on multi-view depth estimation, with our proposed solver consistently improving the depth quality over both conventional and deep learning based MVS pipelines. Code is available at https://github.com/thuzhaowang/idn-solver.
Depth estimation from a single image is an important task that can be applied to various fields in computer vision, and has grown rapidly with the development of convolutional neural networks. In this paper, we propose a novel structure and training strategy for monocular depth estimation to further improve the prediction accuracy of the network. We deploy a hierarchical transformer encoder to capture and convey the global context, and design a lightweight yet powerful decoder to generate an estimated depth map while considering local connectivity. By constructing connected paths between multi-scale local features and the global decoding stream with our proposed selective feature fusion module, the network can integrate both representations and recover fine details. In addition, the proposed decoder shows better performance than the previously proposed decoders, with considerably less computational complexity. Furthermore, we improve the depth-specific augmentation method by utilizing an important observation in depth estimation to enhance the model. Our network achieves state-of-the-art performance over the challenging depth dataset NYU Depth V2. Extensive experiments have been conducted to validate and show the effectiveness of the proposed approach. Finally, our model shows better generalisation ability and robustness than other comparative models.
RGB-D data is essential for solving many problems in computer vision. Hundreds of public RGB-D datasets containing various scenes, such as indoor, outdoor, aerial, driving, and medical, have been proposed. These datasets are useful for different applications and are fundamental for addressing classic computer vision tasks, such as monocular depth estimation. This paper reviewed and categorized image datasets that include depth information. We gathered 203 datasets that contain accessible data and grouped them into three categories: scene/objects, body, and medical. We also provided an overview of the different types of sensors, depth applications, and we examined trends and future directions of the usage and creation of datasets containing depth data, and how they can be applied to investigate the development of generalizable machine learning models in the monocular depth estimation field.
We propose a depth estimation method from a single-shot monocular endoscopic image using Lambertian surface translation by domain adaptation and depth estimation using multi-scale edge loss. We employ a two-step estimation process including Lambertian surface translation from unpaired data and depth estimation. The texture and specular reflection on the surface of an organ reduce the accuracy of depth estimations. We apply Lambertian surface translation to an endoscopic image to remove these texture and reflections. Then, we estimate the depth by using a fully convolutional network (FCN). During the training of the FCN, improvement of the object edge similarity between an estimated image and a ground truth depth image is important for getting better results. We introduced a muti-scale edge loss function to improve the accuracy of depth estimation. We quantitatively evaluated the proposed method using real colonoscopic images. The estimated depth values were proportional to the real depth values. Furthermore, we applied the estimated depth images to automated anatomical location identification of colonoscopic images using a convolutional neural network. The identification accuracy of the network improved from 69.2% to 74.1% by using the estimated depth images.
Recently, self-supervised learning of depth and ego-motion from thermal images shows strong robustness and reliability under challenging scenarios. However, the inherent thermal image properties such as weak contrast, blurry edges, and noise hinder to generate effective self-supervision from thermal images. Therefore, most research relies on additional self-supervision sources such as well-lit RGB images, generative models, and Lidar information. In this paper, we conduct an in-depth analysis of thermal image characteristics that degenerates self-supervision from thermal images. Based on the analysis, we propose an effective thermal image mapping method that significantly increases image information, such as overall structure, contrast, and details, while preserving temporal consistency. The proposed method shows outperformed depth and pose results than previous state-of-the-art networks without leveraging additional RGB guidance.