Yu Chunan

Yu Chunan

Even if the road is long, you will reach your destination. Even if the task is difficult, you will succeed.

Publications

Rapid 3D Model Generation with Intuitive 3D Input

Published in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)., 2024

The paper proposes Deep3DVRSketch, a 3D model generation network that enables novice users to create high-quality 3D models from 3D VR sketches. It aims to overcome the complexity of traditional CAD software. The method uses AI models and introduces the KO3D+ dataset. Experiments show it is over 10 times faster than conventional CAD tools.

Recommended citation: Chen T, Ding C, Zhang S, et al. Rapid 3d model generation with intuitive 3d input[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2024: 12554-12564.
Download Paper

Spatio-Temporal Action Detection with a Motion Sense and Semantic Correction Framework

Published in 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024

The paper proposes a new framework called Motion Sense and Semantic Correction (MS-SC) to address the challenges of distinguishing between action and non-action features and the fusion of information across different modalities in spatio-temporal action detection. The Motion Sense Module (MSM) increases the distance between action and non-action features, improving feature discriminability. The Semantic Correction Fusion Module (SFM) facilitates interaction between features from different modalities, maximizing information integration.

Recommended citation: Zhang Y, Yu C, Fu C, et al. Spatio-Temporal Action Detection with a Motion Sense and Semantic Correction Framework[C]//ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2024: 3645-3649.
Download Paper

Paper Title Number 4

Published in GitHub Journal of Bugs, 2024

This paper is about fixing template issue #693.

Recommended citation: Your Name, You. (2024). "Paper Title Number 3." GitHub Journal of Bugs. 1(3).
Download Paper

Freshness uniformity measurement network based on multi-layer feature fusion and histogram layer

Published in Signal, Image and Video Processing, 2023

This paper introduces the Multi-Scale Feature Histogram Network (MFHisNet) for texture classification, addressing issues in existing algorithms such as inadequate multi-scale feature representation and lack of feature selection.

Recommended citation: Zang Y, Yu C, Fu C, et al. Freshness uniformity measurement network based on multi-layer feature fusion and histogram layer[J]. Signal, Image and Video Processing, 2024, 18(2): 1525-1538.
Download Paper