Lune

Profiles2020-2023

Dong Xu

The University of Hong Kong

13Papers
12Last author
21Unique co-authors
0First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers13

  1. VL-SAT: Visual-Linguistic Semantics Assisted Training for 3D Semantic Scene Graph Prediction in Point CloudAuthor 4 of 6CVPR 2023
  2. 3DJCG: A Unified Framework for Joint Dense Captioning and Visual Grounding on 3D Point CloudsAuthor 5 of 5CVPR 2022 · 80 citations
  3. LSVC: A Learning-based Stereo Video Compression FrameworkAuthor 6 of 6CVPR 2022 · 39 citations
  4. Learning based Multi-modality Image and Video CompressionAuthor 5 of 5CVPR 2022 · 26 citations
  5. 3DVG-Transformer: Relation Modeling for Visual Grounding on Point CloudsAuthor 4 of 4ICCV 2021 · 234 citations
  6. StyleFormer: Real-time Arbitrary Style Transfer via Parametric Style CompositionAuthor 4 of 4ICCV 2021 · 130 citations
  7. STVGBert: A Visual-linguistic Transformer based Framework for Spatio-temporal Video GroundingAuthor 3 of 3ICCV 2021 · 75 citations
  8. Back-Tracing Representative Points for Voting-Based 3D Object Detection in Point CloudsAuthor 5 of 5CVPR 2021
  9. FVC: A New Framework Towards Deep Video Compression in Feature SpaceAuthor 3 of 3CVPR 2021
  10. Inception Convolution With Efficient Dilation SearchAuthor 8 of 8CVPR 2021
  11. VoxelContext-Net: An Octree Based Framework for Point Cloud CompressionAuthor 3 of 3CVPR 2021
  12. Channel Pruning Guided by Classification Loss and Feature ImportanceAuthor 3 of 3AAAI 2020 · 59 citations
  13. Multi-Dimensional Pruning: A Unified Framework for Model CompressionAuthor 3 of 3CVPR 2020
Dusk over the sea between two cliffs drawn in fine vertical lines