arXiv subjects

Wei Yin

Publications and source records attributed to Wei Yin.

88 records · Page 5Linked to original sources

arXiv2022

Exploiting Correspondences with All-pairs Correlations for Multi-view Depth Estimation

Multi-view depth estimation plays a critical role in reconstructing and understanding the 3D world. Recent learning-based methods have made significant progress in it. However, multi-view depth estimation is fundamentally a correspondence-based optimization problem, but previous learning-based methods mainly rely on predefined depth hypotheses to build correspondence as the cost volume and implicitly regularize it to fit depth prediction, deviating from the essence of iterative optimization based on stereo correspondence. Thus, they suffer unsatisfactory precision and generalization capability. In this paper, we are the first to explore more general image correlations to establish correspondences dynamically for depth estimation. We design a novel iterative multi-view depth estimation framework mimicking the optimization process, which consists of 1) a correlation volume construction module that models the pixel similarity between a reference image and source images as all-to-all correlations; 2) a flow-based depth initialization module that estimates the depth from the 2D optical flow; 3) a novel correlation-guided depth refinement module that reprojects points in different views to effectively fetch relevant correlations for further fusion and integrate the fused correlation for iterative depth update. Without predefined depth hypotheses, the fused correlations establish multi-view correspondence in an efficient way and guide the depth refinement heuristically. We conduct sufficient experiments on ScanNet, DeMoN, ETH3D, and 7Scenes to demonstrate the superiority of our method on multi-view depth estimation and its best generalization ability.

cs.CV↗

arXiv2022

Improving Monocular Visual Odometry Using Learned Depth

Monocular visual odometry (VO) is an important task in robotics and computer vision. Thus far, how to build accurate and robust monocular VO systems that can work well in diverse scenarios remains largely unsolved. In this paper, we propose a framework to exploit monocular depth estimation for improving VO. The core of our framework is a monocular depth estimation module with a strong generalization capability for diverse scenes. It consists of two separate working modes to assist the localization and mapping. With a single monocular image input, the depth estimation module predicts a relative depth to help the localization module on improving the accuracy. With a sparse depth map and an RGB image input, the depth estimation module can generate accurate scale-consistent depth for dense mapping. Compared with current learning-based VO methods, our method demonstrates a stronger generalization ability to diverse scenes. More significantly, our framework is able to boost the performances of existing geometry-based VO methods by a large margin.

cs.CV↗

arXiv2022

Pseudo-LiDAR Based Road Detection

Road detection is a critically important task for self-driving cars. By employing LiDAR data, recent works have significantly improved the accuracy of road detection. Relying on LiDAR sensors limits the wide application of those methods when only cameras are available. In this paper, we propose a novel road detection approach with RGB being the only input during inference. Specifically, we exploit pseudo-LiDAR using depth estimation, and propose a feature fusion network where RGB and learned depth information are fused for improved road detection. To further optimize the network structure and improve the efficiency of the network. we search for the network structure of the feature fusion module using NAS techniques. Finally, be aware of that generating pseudo-LiDAR from RGB via depth estimation introduces extra computational costs and relies on depth estimation networks, we design a modality distillation strategy and leverage it to further free our network from these extra computational cost and dependencies during inference. The proposed method achieves state-of-the-art performance on two challenging benchmarks, KITTI and R2D.

cs.CV↗

arXiv2022

Retrieval Augmented Classification for Long-Tail Visual Recognition

We introduce Retrieval Augmented Classification (RAC), a generic approach to augmenting standard image classification pipelines with an explicit retrieval module. RAC consists of a standard base image encoder fused with a parallel retrieval branch that queries a non-parametric external memory of pre-encoded images and associated text snippets. We apply RAC to the problem of long-tail classification and demonstrate a significant improvement over previous state-of-the-art on Places365-LT and iNaturalist-2018 (14.5% and 6.7% respectively), despite using only the training datasets themselves as the external information source. We demonstrate that RAC's retrieval module, without prompting, learns a high level of accuracy on tail classes. This, in turn, frees the base encoder to focus on common classes, and improve its performance thereon. RAC represents an alternative approach to utilizing large, pretrained models without requiring fine-tuning, as well as a first step towards more effectively making use of external memory within common computer vision architectures.

cs.CV↗

arXiv2021

Virtual Normal: Enforcing Geometric Constraints for Accurate and Robust Depth Prediction

Monocular depth prediction plays a crucial role in understanding 3D scene geometry. Although recent methods have achieved impressive progress in terms of evaluation metrics such as the pixel-wise relative error, most methods neglect the geometric constraints in the 3D space. In this work, we show the importance of the high-order 3D geometric constraints for depth prediction. By designing a loss term that enforces a simple geometric constraint, namely, virtual normal directions determined by randomly sampled three points in the reconstructed 3D space, we significantly improve the accuracy and robustness of monocular depth estimation. Significantly, the virtual normal loss can not only improve the performance of learning metric depth, but also disentangle the scale information and enrich the model with better shape information. Therefore, when not having access to absolute metric depth training data, we can use virtual normal to learn a robust affine-invariant depth generated on diverse scenes. In experiments, We show state-of-the-art results of learning metric depth on NYU Depth-V2 and KITTI. From the high-quality predicted depth, we are now able to recover good 3D structures of the scene such as the point cloud and surface normal directly, eliminating the necessity of relying on additional models as was previously done. To demonstrate the excellent generalizability of learning affine-invariant depth on diverse data with the virtual normal loss, we construct a large-scale and diverse dataset for training affine-invariant depth, termed Diverse Scene Depth dataset (DiverseDepth), and test on five datasets with the zero-shot test setting. Code is available at: https://git.io/Depth

cs.CV↗

arXiv2021

Generic Perceptual Loss for Modeling Structured Output Dependencies

The perceptual loss has been widely used as an effective loss term in image synthesis tasks including image super-resolution, and style transfer. It was believed that the success lies in the high-level perceptual feature representations extracted from CNNs pretrained with a large set of images. Here we reveal that, what matters is the network structure instead of the trained weights. Without any learning, the structure of a deep network is sufficient to capture the dependencies between multiple levels of variable statistics using multiple layers of CNNs. This insight removes the requirements of pre-training and a particular network structure (commonly, VGG) that are previously assumed for the perceptual loss, thus enabling a significantly wider range of applications. To this end, we demonstrate that a randomly-weighted deep CNN can be used to model the structured dependencies of outputs. On a few dense per-pixel prediction tasks such as semantic segmentation, depth estimation and instance segmentation, we show improved results of using the extended randomized perceptual loss, compared to the baselines using pixel-wise loss alone. We hope that this simple, extended perceptual loss may serve as a generic structured-output loss that is applicable to most structured output learning tasks.

cs.CV↗

arXiv2020

Learning to Recover 3D Scene Shape from a Single Image

Despite significant progress in monocular depth estimation in the wild, recent state-of-the-art methods cannot be used to recover accurate 3D scene shape due to an unknown depth shift induced by shift-invariant reconstruction losses used in mixed-data depth prediction training, and possible unknown camera focal length. We investigate this problem in detail, and propose a two-stage framework that first predicts depth up to an unknown scale and shift from a single monocular image, and then use 3D point cloud encoders to predict the missing depth shift and focal length that allow us to recover a realistic 3D scene shape. In addition, we propose an image-level normalized regression loss and a normal-based geometry loss to enhance depth prediction models trained on mixed datasets. We test our depth model on nine unseen datasets and achieve state-of-the-art performance on zero-shot dataset generalization. Code is available at: https://git.io/Depth

cs.CV↗

arXiv2020

Hierarchical Text Interaction for Rating Prediction

Traditional recommender systems encounter several challenges such as data sparsity and unexplained recommendation. To address these challenges, many works propose to exploit semantic information from review data. However, these methods have two major limitations in terms of the way to model textual features and capture textual interaction. For textual modeling, they simply concatenate all the reviews of a user/item into a single review. However, feature extraction at word/phrase level can violate the meaning of the original reviews. As for textual interaction, they defer the interactions to the prediction layer, making them fail to capture complex correlations between users and items. To address those limitations, we propose a novel Hierarchical Text Interaction model(HTI) for rating prediction. In HTI, we propose to model low-level word semantics and high-level review representations hierarchically. The hierarchy allows us to exploit textual features at different granularities. To further capture complex user-item interactions, we propose to exploit semantic correlations between each user-item pair at different hierarchies. At word level, we propose an attention mechanism specialized to each user-item pair, and capture the important words for representing each review. At review level, we mutually propagate textual features between the user and item, and capture the informative reviews. The aggregated review representations are integrated into a collaborative filtering framework for rating prediction. Experiments on five real-world datasets demonstrate that HTI outperforms state-of-the-art models by a large margin. Further case studies provide a deep insight into HTI's ability to capture semantic correlations at different levels of granularities for rating prediction.

cs.IR↗

arXiv2020

DiverseDepth: Affine-invariant Depth Prediction Using Diverse Data

We present a method for depth estimation with monocular images, which can predict high-quality depth on diverse scenes up to an affine transformation, thus preserving accurate shapes of a scene. Previous methods that predict metric depth often work well only for a specific scene. In contrast, learning relative depth (information of being closer or further) can enjoy better generalization, with the price of failing to recover the accurate geometric shape of the scene. In this work, we propose a dataset and methods to tackle this dilemma, aiming to predict accurate depth up to an affine transformation with good generalization to diverse scenes. First we construct a large-scale and diverse dataset, termed Diverse Scene Depth dataset (DiverseDepth), which has a broad range of scenes and foreground contents. Compared with previous learning objectives, i.e., learning metric depth or relative depth, we propose to learn the affine-invariant depth using our diverse dataset to ensure both generalization and high-quality geometric shapes of scenes. Furthermore, in order to train the model on the complex dataset effectively, we propose a multi-curriculum learning method. Experiments show that our method outperforms previous methods on 8 datasets by a large margin with the zero-shot test setting, demonstrating the excellent generalization capacity of the learned model to diverse scenes. The reconstructed point clouds with the predicted depth show that our method can recover high-quality 3D shapes. Code and dataset are available at: https://tinyurl.com/DiverseDepth

cs.CV↗

arXiv2019

Task-Aware Monocular Depth Estimation for 3D Object Detection

Monocular depth estimation enables 3D perception from a single 2D image, thus attracting much research attention for years. Almost all methods treat foreground and background regions ("things and stuff") in an image equally. However, not all pixels are equal. Depth of foreground objects plays a crucial role in 3D object recognition and localization. To date how to boost the depth prediction accuracy of foreground objects is rarely discussed. In this paper, we first analyse the data distributions and interaction of foreground and background, then propose the foreground-background separated monocular depth estimation (ForeSeE) method, to estimate the foreground depth and background depth using separate optimization objectives and depth decoders. Our method significantly improves the depth estimation performance on foreground objects. Applying ForeSeE to 3D object detection, we achieve 7.5 AP gains and set new state-of-the-art results among other monocular methods. Code will be available at: https://github.com/WXinlong/ForeSeE.

cs.CV↗

arXiv2019

Auxiliary Learning for Deep Multi-task Learning

Multi-task learning (MTL) is an efficient solution to solve multiple tasks simultaneously in order to get better speed and performance than handling each single-task in turn. The most current methods can be categorized as either: (i) hard parameter sharing where a subset of the parameters is shared among tasks while other parameters are task-specific; or (ii) soft parameter sharing where all parameters are task-specific but they are jointly regularized. Both methods suffer from limitations: the shared hidden layers of the former are difficult to optimize due to the competing objectives while the complexity of the latter grows linearly with the increasing number of tasks. To mitigate those drawbacks, this paper proposes an alternative, where we explicitly construct an auxiliary module to mimic the soft parameter sharing for assisting the optimization of the hard parameter sharing layers in the training phase. In particular, the auxiliary module takes the outputs of the shared hidden layers as inputs and is supervised by the auxiliary task loss. During training, the auxiliary module is jointly optimized with the MTL network, serving as a regularization by introducing an inductive bias to the shared layers. In the testing phase, only the original MTL network is kept. Thus our method avoids the limitation of both categories. We evaluate the proposed auxiliary module on pixel-wise prediction tasks, including semantic segmentation, depth estimation, and surface normal prediction with different network structures. The extensive experiments over various settings verify the effectiveness of our methods.

cs.CV↗

arXiv2019

Enforcing geometric constraints of virtual normal for depth prediction

Monocular depth prediction plays a crucial role in understanding 3D scene geometry. Although recent methods have achieved impressive progress in evaluation metrics such as the pixel-wise relative error, most methods neglect the geometric constraints in the 3D space. In this work, we show the importance of the high-order 3D geometric constraints for depth prediction. By designing a loss term that enforces one simple type of geometric constraints, namely, virtual normal directions determined by randomly sampled three points in the reconstructed 3D space, we can considerably improve the depth prediction accuracy. Significantly, the byproduct of this predicted depth being sufficiently accurate is that we are now able to recover good 3D structures of the scene such as the point cloud and surface normal directly from the depth, eliminating the necessity of training new sub-models as was previously done. Experiments on two benchmarks: NYU Depth-V2 and KITTI demonstrate the effectiveness of our method and state-of-the-art performance.

cs.CV↗

arXiv2019

Temporal phase unwrapping using deep learning

The multi-frequency temporal phase unwrapping (MF-TPU) method, as a classical phase unwrapping algorithm for fringe projection profilometry (FPP), is capable of eliminating the phase ambiguities even in the presence of surface discontinuities or spatially isolated objects. For the simplest and most efficient case, two sets of 3-step phase-shifting fringe patterns are used: the high-frequency one is for 3D measurement and the unit-frequency one is for unwrapping the phase obtained from the high-frequency pattern set. The final measurement precision or sensitivity is determined by the number of fringes used within the high-frequency pattern, under the precondition that the phase can be successfully unwrapped without triggering the fringe order error. Consequently, in order to guarantee a reasonable unwrapping success rate, the fringe number (or period number) of the high-frequency fringe patterns is generally restricted to about 16, resulting in limited measurement accuracy. On the other hand, using additional intermediate sets of fringe patterns can unwrap the phase with higher frequency, but at the expense of a prolonged pattern sequence. Inspired by recent successes of deep learning techniques for computer vision and computational imaging, in this work, we report that the deep neural networks can learn to perform TPU after appropriate training, as called deep-learning based temporal phase unwrapping (DL-TPU), which can substantially improve the unwrapping reliability compared with MF-TPU even in the presence of different types of error sources, e.g., intensity noise, low fringe modulation, and projector nonlinearity. We further experimentally demonstrate for the first time, to our knowledge, that the high-frequency phase obtained from 64-period 3-step phase-shifting fringe patterns can be directly and reliably unwrapped from one unit-frequency phase using DL-TPU.

eess.IV↗

arXiv2018

MAC-Layer Rate Control for 802.11 Networks: Lesson Learned and Looking Forward

Rate control at the MAC-layer is one of the fundamental building blocks in many wireless networks. Over the past two decades around thirty mechanisms have been proposed in the literature. Among them, there are mechanisms that make rate selection decisions based on sophisticated measurements of wireless link quality, and others that are based on straight-forward heuristics. Minstrel, for example, is an elegant mechanism that has been adopted by hundreds of millions of computers, yet, not much was known about its performance until recently. The purpose of this paper is to provide a comprehensive survey and analysis of existing solutions from the two fundamental aspects of rate control - metrics and algorithms. We also review how these solutions were evaluated and compared against each other. Based on our detailed studies and observations, we share important insights on future development of rate control mechanisms at the MAC-layer. This discussion also takes into account the recent developments in wireless technologies and emerging applications, such as Internet-of-Things, and shows issues that need to be addressed in the design of new rate control mechanisms suitable for these technologies and applications.

cs.NI↗

arXiv2018

Fringe pattern analysis using deep learning

In many optical metrology techniques, fringe pattern analysis is the central algorithm for recovering the underlying phase distribution from the recorded fringe patterns. Despite extensive research efforts for decades, how to extract the desired phase information, with the highest possible accuracy, from the minimum number of fringe patterns remains one of the most challenging open problems. Inspired by recent successes of deep learning techniques for computer vision and other applications, here, we demonstrate for the first time, to our knowledge, that the deep neural networks can be trained to perform fringe analysis, which substantially enhances the accuracy of phase demodulation from a single fringe pattern. The effectiveness of the proposed method is experimentally verified using carrier fringe patterns under the scenario of fringe projection profilometry. Experimental results demonstrate its superior performance in terms of high accuracy and edge-preserving over two representative single-frame techniques: Fourier transform profilometry and Windowed Fourier profilometry.

eess.IV↗

arXiv2016

In Situ Simultaneous Measuring Forces Acted on Individual Legs of Water Striders According to Archimedes' Principle

The supporting forces of all the legs have been simultaneously measured through a natural phenomenon inspired shadow method, by capturing water strider leg shadows to calculate the expelled water volume and the equivalent floating forces according to Archimedes' principle, with a sensitivity up to nN~pN/pixel. The method was verified by comparing to electronic balance to weight water striders. The supporting force acted on legs was found to linearly proportional to its shadow area, which geometry clearly indicated the hydrophobicity of legs. The pressed depth of legs, vertical weight focus position change and body pitch angle during leg refreshing, sculling and jumping forward of water striders have been achieved with resolutions of 5um/pixel, 2 um/pixel, and 0.02 degree, respectively, which is difficult for general imaging technologies. The shadow method was also developed into a force measuring device and characterized the adhesion force of a hair/polymer surface contact in a few uN with a 0.3 nN/pixel resolution.

cond-mat.soft↗