Tại hội thảo CVPR năm nay, Facebook AI đã mang công nghệ thuộc các lĩnh vực trong Thị giác máy tính (Computer vision – CV) lên một tầm cao mới, trong đó bao gồm các tác vụ phân khúc chính, tìm kiếm kiến trúc, học chuyển giao, và học đa mô hình. Bài viết sẽ chia sẻ toàn bộ các tóm tắt và chi tiết về những nghiên cứu của Facebook Research được trình bày tại CVPR.

1. 12-in-1: Multitask vision and language representation learning
Jiasen Lu, Vedanuj Goswami, Marcus Rohrbach, Devi Parikh, Stefan Lee

2. A multigrid method for efficiently training video models
Chao-Yuan Wu, Ross Girshick, Kaiming He, Christoph Feichtenhofer, Philipp Krähenbühl

3. ARCH: Animatable Reconstruction of Clothed Humans
Zeng Huang, Yuanlu Xu, Christoph Lassner, Hao Li, Tony Tung

4. Articulation-aware Canonical Surface Mapping
Nilesh Kulkarni, Abhinav Gupta, David F. Fouhey, Shubham Tulsiani

5. Classifying, segmenting, and tracking object instances in video with mask propagation
Gedas Bertasius, Lorenzo Torresani

6. Cluster and relearn: Improving generalization of visual representations
Xueting Yan, Ishan Misra, Abhinav Gupta, Deepti Ghadiyaram, Dhruv Mahajan

7. Designing network design spaces
Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, Piotr Dollar

8. Don’t judge an object by its context: Learning to overcome contextual bias
Krishna Kumar Singh, Dhruv Mahajan, Kristen Grauman, Yong Jae Lee, Matt Feiszli, Deepti Ghadiyaram

9. EGO-TOPO: Environment affordances from egocentric video
Tushar Nagarajan, Yanghao Li, Christoph Feichtenhofer, Kristen Grauman

10. End-to-end view synthesis from a single image
Olivia Wiles, Georgia Gkioxari, Richard Szeliski, Justin Johnson

11. Epipolar transformers
Yihui He, Rui Yan, Katerina Fragkiadaki, Shoou-I Yu

12. FBNetV2: Differentiable neural architecture search for spatial and channel cimensionss
Alvin Wan, Xiaoliang Dai, Peizhao Zhang, Zijian He, Yuandong Tian, Saining Xie, Bichen Wu, Matthew Yu, Tao Xu, Kan Chen, Peter Vajda, Joseph E. Gonzalez

13. From Paris to Berlin: Discovering style influences around the world
Ziad Al-Halah,Kristen Grauman

14. From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality
Zhenqiang Ying, Haoran Niu, Praful Gupta, Dhruv Mahajan, Deepti Ghadiyaram, Alan Bovik

15. GrappaNet: Combining Parallel Imaging with Deep Learning for Multi-Coil MRI Reconstruction

Anuroop Sriram, Jure Zbontar, Tullie Murrell, C. Lawrence Zitnick, Aaron Defazio, Daniel K. Sodickson

16. Hierarchical scene coordinate classification and regression for visual localization
Xiaotian Li, Shuzhe Wang, Yi Zhao,Jakob Verbeek, Juho Kannala

17. Improving lowshot object detection with weakly labeled Data
Vignesh Ramanathan, Rui Wang, Dhruv Mahajan

18. ImVoteNet: Boosting 3D object detection in point clouds with image votes
Charles R. Qi, Xinlei Chen, Or Litany, Leonidas J. Guibas

19. In defense of grid features for visual question answering
Huaizu Jiang, Ishan Misra, Marcus Rohrbach, Erik Learned-Miller, Xinlei Chen

20. Iterative answer prediction with pointer-augmented multimodal transformers for TextVQA
Ronghang Hu, Amanpreet Singh, Trevor Darrell, Marcus Rohrbach

(Còn nữa)

Theo Facebook AI

Tin liên quan: