Robotics and Computer Vision Lab

Publications

International Conference
2020.07.03 13:52

Non-Local Spatial Propagation Network for Depth Completion

조회 수 689 댓글 0
Extra Form
저 자 Jinsun Park, Kyungdon Joo, Zhe Hu, Chi-Kuei Liu, In So Kweon
학 회 European Conference on Computer Vision (ECCV)
논문일시(Year) 2020
논문일시(Month) 08

In this paper, we propose a robust and efficient end-to-end non-local spatial propagation network for depth completion. The proposed network takes RGB and sparse depth images as inputs and estimates non-local neighbors and their affinities of each pixel, as well as an initial depth map with pixel-wise confidences. The initial depth prediction is then iteratively refined by its confidence and non-local spatial propagation procedure based on the predicted non-local neighbors and corresponding affinities. Unlike previous algorithms that utilize fixed-local neighbors, the proposed algorithm effectively avoids irrelevant local neighbors and concentrates on relevant non-local neighbors during propagation. In addition, we introduce a learnable affinity normalization to better learn the affinity combinations compared to conventional methods. The proposed algorithm is inherently robust to the mixed-depth problem on depth boundaries, which is one of the major issues for existing depth estimation/completion algorithms. Experimental results on indoor and outdoor datasets demonstrate that the proposed algorithm is superior to conventional algorithms in terms of depth completion accuracy and robustness to the mixed-depth problem. Our implementation is publicly available on the project page.

 

Acknowledgement :

This work was partially supported by the National Information Society Agency for construction of training data for artificial intelligence (2100-2131-305-107-19).

 


List of Articles
620. Learning to Associate Every Segment for Video Panoptic Segmentation
Sanghyun Woo, Dahun Kim, Joon-Young Lee and In So Kweon
Computer Vision and Pattern Recognition, CVPR, 2021 2021 / 06
619. Depth Completion using Plane-Residual Representation
Byeong-Uk Lee, Kyunghyun Lee and In So Kweon
Computer Vision and Pattern Recognition, CVPR, 2021 2021 / 06
618. Stereo Object Matching Network
{Jaesung Choe, Kyungdon Joo}*, Francois Rameau, and In So Kweon
IEEE International Conference on Robotics and Automation (ICRA) 2021 / 06
617. Volumetric Propagation Network: Stereo-LiDAR Fusion for Long Range Depth Estimation
Jaesung Choe, Kyungdon Joo, Imtiaz Tooba, In So Kweon
IEEE Robotics and Automation Letters (RA-L) 2021 / 06
616. Universal Adversarial Perturbations Through the Lens of Deep Steganography: Towards A Fourier Perspective
{Chaoning Zhang, Philipp Benz}*, Adil Karjauv, In So Kweon
Association for the Advancement of Artificial Intelligence (AAAI) 2021 / 02
615. Optical Flow Estimation from a Single Motion-blurred Image
Dawit Mureja Argaw, Junsik Kim, Francois Rameau, Jae Won Cho, In So Kweon
Association for the Advancement of Artificial Intelligence (AAAI) 2021 / 02
614. Motion-blurred Video Interpolation and Extrapolation
Dawit Mureja Argaw, Junsik Kim, Francois Rameau, In So Kweon
Association for the Advancement of Artificial Intelligence (AAAI) 2021 / 02
613. Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection Consistency
Seokju Lee, Sunghoon Im, Stephen Lin, In So Kweon
Association for the Advancement of Artificial Intelligence (AAAI) 2021 / 02
612. ResNet or DenseNet? Introducing Dense Shortcuts to ResNet
Chaoning Zhang*, Philipp Benz*, Dawit Mureja Argaw, Seokju Lee, Junsik Kim, Francois Rameau, Jean-Charles Bazin, In So Kweon (*: equal contribution)
IEEE Winter Conference on Applications of Computer Vision (WACV) 2021 / 1
611. The Devil is in the Boundary: Exploiting Boundary Representation for Basis-based Instance Segmentation
Myungchul Kim, Sanghyun Woo, Dahun Kim, and In So Kweon
IEEE Winter Conference on Applications of Computer Vision (WACV) 2021 / 01
610. Revisiting Batch Normalization for Improving Corruption Robustness
Philipp Benz*, Chaoning Zhang*, Adil Karjauv, and In So Kweon (*: equal contribution)
IEEE Winter Conference on Applications of Computer Vision (WACV) 2021 / 01
609. High-quality Frame Interpolation via Tridirectional Inference
Jinsoo Choi, Jaesik Park, and In So Kweon
IEEE Winter Conference on Applications of Computer Vision (WACV) 2021 / 01
608. UDH: Universal Deep Hiding for Steganography, Watermarking, and Light Field Messaging
Chaoning Zhang*, Philipp Benz*, Adil Karjauv*, Geng Sun, In-So Kweon (*: equal-contribution)
NeurIPS, 2020 2020 / 12
607. Discover, Hallucinate, and Adapt: Open Compound Domain Adaptation for Semantic Segmentation
KwanYong Park, Sanghyun Woo, Inkyu Shin, In So Kweon
NeurIPS, 2020 2020 / 12
606. An Efficient Asynchronous Method for Integrating Evolutionary and Gradient-based Policy Search
Kyunghyun Lee, Byeong-Uk Lee, Ukcheol Shin, In So Kweon
NeurIPS, 2020 2020 / 12
605. Align-and-Attend Network for Globally and Locally Coherent Video Inpainting
Sanghyun Woo, Dahun Kim, KwanYong Park, Joon-Young Lee, In So Kweon
British Machine Vision Conference (BMVC) 2020 / 09
604. SideGuide: A Large-scale Sidewalk Dataset for Guiding Impaired People
Kibaek Park*, Youngtaek Oh*, Soomin Ham*, Kyungdon Joo*, HYOKYOUNG KIM, HyoYoung Kum, In So Kweon
IROS, 2020 2020 / 10
603. Two-Phase Pseudo Label Densification for Self-training based Domain Adaptation
Inkyu Shin, Sanghyun Woo, Fei Pan, In So Kweon
European Conference on Computer Vision (ECCV) 2020 / 08
602. Global-and-Local Relative Position Embedding for Unsupervised Video Summarization
Yunjae Jung, Donghyeon Cho, Sanghyun Woo, In So Kweon
European Conference on Computer Vision (ECCV) 2020 / 08
» Non-Local Spatial Propagation Network for Depth Completion
Jinsun Park, Kyungdon Joo, Zhe Hu, Chi-Kuei Liu, In So Kweon
European Conference on Computer Vision (ECCV) 2020 / 08
Board Pagination Prev 1 2 3 4 5 6 7 8 9 10 ... 31 Next
/ 31