Picture for Hongkai Zhang

Hongkai Zhang

1st Place Solutions of Waymo Open Dataset Challenge 2020 -- 2D Object Detection Track

Add code
Aug 04, 2020
Viaarxiv icon

Appearance-Preserving 3D Convolution for Video-based Person Re-identification

Add code
Jul 27, 2020
Figure 1 for Appearance-Preserving 3D Convolution for Video-based Person Re-identification
Figure 2 for Appearance-Preserving 3D Convolution for Video-based Person Re-identification
Figure 3 for Appearance-Preserving 3D Convolution for Video-based Person Re-identification
Figure 4 for Appearance-Preserving 3D Convolution for Video-based Person Re-identification
Viaarxiv icon

Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training

Add code
Apr 13, 2020
Figure 1 for Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
Figure 2 for Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
Figure 3 for Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
Figure 4 for Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
Viaarxiv icon

Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection

Add code
Jul 16, 2019
Figure 1 for Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection
Figure 2 for Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection
Figure 3 for Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection
Figure 4 for Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection
Viaarxiv icon

WIDER Face and Pedestrian Challenge 2018: Methods and Results

Add code
Feb 19, 2019
Figure 1 for WIDER Face and Pedestrian Challenge 2018: Methods and Results
Figure 2 for WIDER Face and Pedestrian Challenge 2018: Methods and Results
Figure 3 for WIDER Face and Pedestrian Challenge 2018: Methods and Results
Figure 4 for WIDER Face and Pedestrian Challenge 2018: Methods and Results
Viaarxiv icon