Posts by Collection
portfolio
Me and Sakuma san in Sensetime Japan
After we finished the CVPR submission. 
publications
Digging Into Normal Incorporated Stereo Matching
Published in Proceedings of the 30th ACM International Conference on Multimedia.(ACMMM22), 2022
Zihua Liu, Songyan Zhang, Zhicheng Wang and Masatoshi Okutomi
Download here
Global Occlusion-Aware Transformer for Robust Stereo Matching
Published in Proceedings of IEEE/CVF Winter Conference on Applications of Computer Vision(WACV2024), 2023
Zihua Liu, Yizhou Li and Masatoshi Okutomi
Download here
VSRD: Instance-Aware Volumetric Silhouette Rendering for Weakly Supervised 3D Object Detection
Published in IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR2024), 2024
Zihua Liu, Hiroki Sakuma and Masatoshi Okutomi
Download here
CFDNet: A Generalizable Foggy Stereo Matching Network with Contrastive Feature Distillation
Published in IEEE International Conference on Robotics and Automation (ICRA2024), 2024
Zihua Liu, Yizhou Li and Masatoshi Okutomi
Download here
talks
Presentation on ACM MM22
Published:
Presentation on ICRA 2024
Published:
Presentation on WACV 2024
Published:
teaching
MEGVII Algorthim Engineer Internship
Undergraduate course, MEGVII Shanghai Reseach Center, 2022
- Focus on Parking Slot detection and segmentation for autonomous drving.
- Add a angle constrain to the corners of the parking slot which improve the recall rate of parking slot detection by 10%.
- Impose a novel training strategy for data augmentation which make full use of the limited training data, which greatly improve the detection at ill-position like image edges and occluded regions. It was finally accepted by Megvii as the IPM pipeline for automatic parking algorithm.
Sensetime Japan, Researcher Intren
Workshop, Sensetime Tokyo Office, 2023
- Conduct research on weakly-supervised 3D detections without using LiDAR point clouds or annotations.
- Write academic papers.