Measuring the Effects of Temporal Coherence in Depth Estimation for Dynamic Scenes

Self-supervised depth estimation methods can achieve competitive performance using only unlabeled monocular videos, but they suffer from the uncertainty of jointly learning depth and pose without any ground truths of both tasks. Supervised framework provides robust and superior performance but is limited by the scope of the labeled data. In this paper, we introduce SENSE, a novel learning paradigm for self-supervised monocular depth estimation that progressively evolves the prediction result using supervised learning, but without requiring labeled data. The key contribution of our approach stems from the novel use of the pseudo labels - the noisy depth estimation from the self-supervised methods. We surprisingly find that a fully supervised depth estimation network trained using the pseudo labels can produce even better results than its "ground truth". To push the envelope further, we then evolve the self-supervised backbone by replacing its depth estimation branch with that fully supervised network. Based on this idea, we devise a comprehensive training pipeline that alternatively enhances the two key branches (depth and pose estimation) of the self-supervised backbone network. Our proposed approach can effectively ease the difficulty of multi-task training in self-supervised depth estimation. Experimental results have shown that our proposed approach achieves state-of-the-art results on the KITTI dataset.

The compression of multiview video in an end-to-end 3D system is required to reduce the amount of visual information. Since multiple cameras usually have a common field of view, high compression ratios can be achieved if both the temporal and inter-view redundancy are exploited. View synthesis prediction is a new coding tool for multiview video that essentially generates virtual views of a scene using images from neighboring cameras and estimated depth values. In this work, we consider depth estimation for view synthesis in multiview video encoding. We focus on generating smooth and accurate depth maps, which can be efficiently coded. We present several improvements to the reference block-based depth estimation approach and demonstrate that the proposed method of depth estimation is not only efficient for view synthesis prediction, but also produces depth maps that require much fewer bits to code.

Methods based on light field information have shown promising results in depth estimation and underwater image restoration. However, improvements are still needed in terms of depth estimation accuracy and image restoration quality. Previous work on underwater image restoration employed an image formation model (IFM) that overlooked the effects of light attenuation and scattering coefficients in underwater environments, leading to unavoidable color deviation and distortion in the restored images. Additionally, the high blurriness and associated distortions in underwater images make depth information extraction and estimation very challenging. In this paper, we refine the light propagation model and propose a method to estimate the attenuation and backscattering coefficients of the underwater IFM. We simplify these coefficients into distance-related functions and design a relationship between distance and the darkest channel to estimate the water coefficients, effectively suppressing color deviation and distortion in the restoration results. Furthermore, to increase the accuracy of depth estimation, we propose using blur cues to construct a cost for refocusing in the depth direction, reducing the impact of high signal-to-noise ratio environments on depth information extraction, and effectively enhancing the accuracy and robustness of depth estimation. Finally, experimental comparisons show that our method achieves more accurate depth estimation and image restoration closer to real scenes compared to state-of-the-art methods.

In this research paper, a system capable of taking as input multiple 2-dimensional images of the surrounding environment from a particular position and creating a 3-dimensional model from them, with navigation possible inside it, for the 360 degree view is developed. Existing approaches for image stitching, which use SIFT features, along with approaches for depth estimation, which use supervised learning to train a Markov Random Field (MRF), are modified in this paper in order to improve their efficiency. Also, an improvement in accuracy of depth estimation is suggested for the 3-dimensional model using matching SIFT features from the multiple input images. A method for enabling navigation in the 3D model through which we prevent motion in areas where movement is not possible is outlined, thus making the 3-dimensional model realistic and suitable for practical use. The proposed system is also an application of Neural Networks.

In this paper, we propose MINE to perform novel view synthesis and depth estimation via dense 3D reconstruction from a single image. Our approach is a continuous depth generalization of the Multiplane Images (MPI) by introducing the NEural radiance fields (NeRF). Given a single image as input, MINE predicts a 4-channel image (RGB and volume density) at arbitrary depth values to jointly reconstruct the camera frustum and fill in occluded contents. The reconstructed and inpainted frustum can then be easily rendered into novel RGB or depth views using differentiable rendering. Extensive experiments on RealEstate10K, KITTI and Flowers Light Fields show that our MINE outperforms state-of-the-art by a large margin in novel view synthesis. We also achieve competitive results in depth estimation on iBims-1 and NYU-v2 without annotated depth supervision. Our source code is available at https://github.com/vincentfung13/MINE.

In this paper, we study a practical omnidirectional depth estimation with neural networks that enables effective learning on real world data obtained using wide-baseline multiple fish-eye cameras. Most previous approaches only used synthetic data providing dense and accurate depth ground truth (GT). However, it is unrealistic to acquire such high quality GT data in real world due to limitations of the existing depth sensors. We first introduce two critical problems that can reduce the accuracy of depth estimation: depth GT sparsity and sensor calibration error. We then propose a novel semi-supervised learning method using pixel-level loss that selectively uses supervised loss and unsupervised re-projection loss according to existence of GT. Empirical results demonstrate that our method efficiently reduces the performance degradation in both simulation on synthetic data and real world data using sparse depth sensor.

Unsupervised deep learning methods have demonstrated an impressive performance for understanding the structure of 3D scene from videos. These data-based learning methods are able to learn the tasks, such as depth, ego-motion, and optical flow estimation. In this paper, we propose a novel unsupervised deep learning method to jointly estimate scene depth, camera ego-motion, and optical flow from stereo images. Consecutive stereo images are used to train the system. After training stage, the system is able to estimate dense depth map, camera 6D pose, and optical flow by using a sequence of monocular images. No labelled data set is required for training. The supervision signals for training three deep neural networks of the system come from various forms of image warping. Due to the use of optical flow, the impact caused by occlusions and moving objects on the estimation results is alleviated. Experiments on the KITTI and Cityscapes datasets show that the proposed system demonstrates a better performance in terms of accuracy in depth, ego-motion, and optical flow estimation.

Robust depth estimation on real-world light field images using Gaussian belief propagation

Firstly, the elementary principle of passive estimation of target depth is studied, which is supported by underwater acoustic detection system of vertical linear array and is conducted in deep sea. Then, error sources of passive estimation of target depth are analysed theoretically from points of view of random error and deployment deviation of vertical linear array, and correlative formulae are deduced. After simulation experiments, it is concluded that how these error sources impact accuracy of passive estimation of target depth, which is the basis of further research for passive estimation of target depth and improvement of accuracy of passive estimation of target depth.

Monocular depth estimation is a fundamental yet challenging task in computer vision as depth information will be lost when 3D scenes are mapped to 2D images. Although deep learning-based methods have led to considerable improvements for this task in a single image, most existing approaches still fail to overcome this limitation. Supervised learning methods model depth estimation as a regression problem and, as a result, require large amounts of ground truth depth data for training in actual scenarios. Unsupervised learning methods treat depth estimation as the synthesis of a new disparity map, which means that rectified stereo image pairs need to be used as the training dataset. Aiming to solve such problem, we present an encoder-decoder based framework, which infers depth maps from monocular video snippets in an unsupervised manner. First, we design an unsupervised learning scheme for the monocular depth estimation task based on the basic principles of structure from motion (SfM) and it only uses adjacent video clips rather than paired training data as supervision. Second, our method predicts two confidence masks to improve the robustness of the depth estimation model to avoid the occlusion problem. Finally, we leverage the largest scale and minimum depth loss instead of the multiscale and average loss to improve the accuracy of depth estimation. The experimental results on the benchmark KITTI dataset for depth estimation show that our method outperforms competing unsupervised methods.

Video depth estimation aims to infer temporally consistent depth. One approach is to finetune a single-image model on each video with geometry constraints, which proves inefficient and lacks robustness. An alternative is learning to enforce consistency from data, which requires well-designed models and sufficient video depth data. To address both challenges, we introduce NVDS that stabilizes inconsistent depth estimated by various single-image models in a plug-and-play manner. We also elaborate a large-scale Video Depth in the Wild (VDW) dataset, which contains 14,203 videos with over two million frames, making it the largest natural-scene video depth dataset. Additionally, a bidirectional inference strategy is designed to improve consistency by adaptively fusing forward and backward predictions. We instantiate a model family ranging from small to large scales for different applications. The method is evaluated on VDW dataset and three public benchmarks. To further prove the versatility, we extend NVDS to video semantic segmentation and several downstream applications like bokeh rendering, novel view synthesis, and 3D reconstruction. Experimental results show that our method achieves significant improvements in consistency, accuracy, and efficiency. Our work serves as a solid baseline and data foundation for learning-based video depth estimation.

The benefits of binocular vision have been debated throughout the history of vision science yet few studies have considered its contribution beyond a viewing distance of a few meters. In the first set of experiments, we compared monocular and binocular performance on depth interval estimation and discrimination tasks at 4.5, 9.0 or 18.0 m. Under monocular conditions, perceived depth was significantly compressed. Binocular depth estimates were much nearer to veridical although also compressed. Regression-based precision measures were much more precise for binocular compared to monocular conditions (ratios between 2.1 and 48). We confirm that stereopsis supports reliable depth discriminations beyond typical laboratory distances. Furthermore, binocular vision can significantly improve both the accuracy and precision of depth estimation to at least 18 m. In another experiment, we used a novel paradigm that allowed the presentation of real binocular disparity stimuli in the presence of rich environmental cues to distance but not interstimulus depth. We found that the presence of environmental cues to distance greatly enhanced stereoscopic depth constancy at distances of 4.5 and 9.0 m. We conclude that stereopsis is an effective cue for depth discrimination and estimation for distances beyond those traditionally assumed. In normal environments, distance information from other sources such as perspective can be effective in scaling depth from disparity.

Depth estimation is an essential component in understanding the 3D geometry of a scene. In comparison to traditional depth estimation methods such as structure from motion and stereo vision matching, determining depth relation using a single camera is challenging. The recent advancements in convolutional neural networks have accelerated the research in monocular depth estimation. However, most technologies infer depth maps using lower resolution images due to network capacity and complexity issues. Another challenge in depth estimation is ambiguous and sparse depth maps. These issues are caused due to labeling errors, hardware faults, or occlusions. This paper presents a novel end-to-end trainable convolutional neural network architecture – depth transverse transformer network (DTTNet). The proposed network is designed and optimized to perform monocular depth estimation. This network aims at exploring the multi-resolution representation to perform pixel-wise depth estimation more accurately. In order to improve the accuracy of depth estimation, different kinds of ad hoc networks are proposed subsequently. Extensive computer simulations on NYU Depth V2 and SUN RGB-D dataset demonstrate the effectiveness of the proposed DTTNet against state-of-the-art methods. DTTNet can potentially optimize depth perception in intelligent systems such as automated driving and video surveillance applications, computational photography, and augmented reality. The source code is available at https://github.com/shreyaskamathkm/DTTNet

Monocular endoscopy images depth estimation with multi-scale residual fusion

View synthesis using depth maps is a crucial application for Free-viewpoint TV (FTV). In this paper, we propose a novel reliability based view synthesis method using two references and their depth maps. The depth estimation with stereo matching is known to be error-prone, leading to noticeable artifacts in the synthesized new views. In order to provide plausible virtual views for FTV, our focus is on the error suppression for the synthesized view. We innovatively introduce the continuous reliability using error approximation by the reference cross-check. The new view interpolation algorithm is generated with the criterion of Least Sum of Squared Errors (LSSE). Furthermore, the proposed algorithm can be considered as a reliable version of the conventional linear view blending. We experimentally demonstrate the effectiveness of our framework with MPEG standard test sequences. The results show that our method outperforms state-of-the-art view interpolation methods both at eliminating artifacts and improving PSNR.