paper-with-me

홈 › Papers

Geometry-Aware Unsupervised Domain Adaptation for Stereo Matching

2021-03-26 · Hiroki Sakuma, Yoshinori Konishi

Recently proposed DNN-based stereo matching methods that learn priors directly from data are known to suffer a drastic drop in accuracy in new environments. Although supervised approaches with ground truth disparity maps often work well, collecting them in each deployment environment is cumbersome and costly. For this reason, many unsupervised domain adaptation methods based on image-to-image translation have been proposed, but these methods do not preserve the geometric structure of a stereo image pair because the image-to-image translation is applied to each view separately. To address this problem, in this paper, we propose an attention mechanism that aggregates features in the left and right views, called Stereoscopic Cross Attention (SCA). Incorporating SCA to an image-to-image translation network makes it possible to preserve the geometric structure of a stereo image pair in the process of the image-to-image translation. We empirically demonstrate the effectiveness of the proposed unsupervised domain adaptation based on the image-to-image translation with SCA.

📄 PDF Abstract BibTeX arXiv:2103.14333

Code (0)

등록된 구현이 없습니다.

Tasks

Domain AdaptationImage-to-Image TranslationStereo MatchingTranslationUnsupervised Domain Adaptation

Similar Papers 제목 키워드 기반

Geometry-based Occlusion-Aware Unsupervised Stereo Matching for Autonomous Driving

2020-10-21 · Liang Peng, Dan Deng, Deng Cai

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. Occlusi…

Autonomous DrivingOcclusion HandlingStereo Matching

RobIA: Robust Instance-aware Continual Test-time Adaptation for Deep Stereo

2025-11-13 · Jueun Ko, Hyewon Park, Hyesong Choi, Dongbo Min arxiv

Stereo Depth Estimation in real-world environments poses significant challenges due to dynamic domain shifts, sparse or unreliable supervision, and the high cost of acquiring dense ground-truth labels. While recent Test-…

Computational EfficiencyStereo Depth EstimationTest-time Adaptation

Learning to Adapt for Stereo

2019-04-05 · CVPR 2019 6 · Alessio Tonioni, Oscar Rahnama, Thomas Joy, Luigi Di Stefano 외

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 …

Autonomous DrivingDepth EstimationStereo Depth Estimation

StereoFlowGAN: Co-training for Stereo and Flow with Unsupervised Domain Adaptation

2023-09-04 · Zhexiao Xiong, Feng Qiao, Yu Zhang, Nathan Jacobs

We introduce a novel training strategy for stereo matching and optical flow estimation that utilizes image-to-image translation between synthetic and real image domains. Our approach enables the training of models that e…

Domain AdaptationImage-to-Image TranslationOptical Flow EstimationStereo Matching+2

Noise-Aware Unsupervised Deep Lidar-Stereo Fusion

2019-04-08 · CVPR 2019 6 · Xuelian Cheng, Yiran Zhong, Yuchao Dao, Pan Ji 외

In this paper, we present LidarStereoNet, the first unsupervised Lidar-stereo fusion network, which can be trained in an end-to-end manner without the need of ground truth depth maps. By introducing a novel "Feedback Loo…

3D geometryDepth CompletionStereo MatchingStereo Matching Hand