Demo Data from EDBT 2026 / reviewers in the wild / expert
Bin Hu 0020
dblp:00/6381-20
· DBLP profile ↗
5ranked-venue papers
4as first author
5since 2021 · last 2026
0000-0002-9925-2552ORCID · conflict
Domains — the database's venue-derived domains; a paper can count in several
Artificial intelligence and machine learning · 3 · 2 first-author · 3 since 2021Graphics, computer vision, multimedia, augmented reality and games · 2 · 1 first-author · 2 since 2021Applied, interdisciplinary, general and emerging computing · 1 · 1 first-author · 1 since 2021
Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.
| Artificial intelligence
2 papers |
Segmentation and scene understanding · 59% Face, body and person analysis · 18% Representation and self-supervised learning · 18% |
Topics — the 7 heaviest of 8, each with the papers that count most for it
| Topic | Weight | Papers | Last | Evidence papers |
|---|---|---|---|---|
Computer vision › Segmentation and scene understanding › object segmentation
class-agnostic segmentation |
0.5 | 1 | 2021 | Weakly-Supervised Instance Segmentation via Class-Agnostic Learning With Salient Images · CVPR 2021 |
Machine learning › Representation and self-supervised learning › representation learning › feature extraction
deep feature learning |
0.5 | 1 | 2021 | Learning generalizable deep feature using triplet-batch-center loss for person re-identification · Sci. China Inf. Sci. 2021 |
Computer vision › Segmentation and scene understanding
instance segmentation |
0.5 | 1 | 2021 | Weakly-Supervised Instance Segmentation via Class-Agnostic Learning With Salient Images · CVPR 2021 |
Computer vision › Face, body and person analysis
person re-identification |
0.5 | 1 | 2021 | Learning generalizable deep feature using triplet-batch-center loss for person re-identification · Sci. China Inf. Sci. 2021 |
Computer vision › Segmentation and scene understanding › instance segmentation
weakly supervised instance segmentation |
0.5 | 1 | 2021 | Weakly-Supervised Instance Segmentation via Class-Agnostic Learning With Salient Images · CVPR 2021 |
Machine learning › Deep learning architectures and training
loss function design |
0.1 | 1 | 2021 | Learning generalizable deep feature using triplet-batch-center loss for person re-identification · Sci. China Inf. Sci. 2021 |
Computer vision › Segmentation and scene understanding › saliency detection
salient object detection |
0.1 | 1 | 2021 | Weakly-Supervised Instance Segmentation via Class-Agnostic Learning With Salient Images · CVPR 2021 |
Methods — techniques the papers use, named apart from their topics
triplet loss · 0.5multi-task learning · 0.5merged and dropped strategy · 0.5box-supervised learning · 0.5batch center loss · 0.5
| Year | Publication | Venue | Position |
|---|---|---|---|
| 2026 | Better early detector for high-performance detection transformer
Bin Hu 0020, Bencheng Liao, Jiyang Qi, Shusheng Yang, Wenyu Liu 0001 |
Image Vis. Comput. | 1 |
| 2025 | Personvit: large-scale self-supervised vision transformer for person re-identification
Bin Hu 0020, Xinggang Wang, Wenyu Liu 0001 |
Mach. Vis. Appl. | 1 |
| 2022 | Joint weakly and fully supervised learning for surface defect segmentation from images
Bin Hu 0020, Xinggang Wang, Wenyong Yu |
Signal Process. Image Commun. | 1 |
| 2021 | Weakly-Supervised Instance Segmentation via Class-Agnostic Learning With Salient ImagesabstractHumans have a strong class-agnostic object segmentation ability and can outline boundaries of unknown objects precisely, which motivates us to propose a box-supervised class-agnostic object segmentation (BoxCaseg) based solution for weakly-supervised instance segmentation. The BoxCaseg model is jointly trained using box-supervised images and salient images in a multi-task learning manner. The fine-annotated salient images provide class-agnostic and precise object localization guidance for box-supervised images. The object masks predicted by a pretrained BoxCaseg model are refined via a novel merged and dropped strategy as proxy ground truth to train a Mask R-CNN for weakly-supervised instance segmentation. Only using 7991 salient images, the weakly-supervised Mask R-CNN is on par with fully-supervised Mask R-CNN on PASCAL VOC and significantly outperforms previous state-of-the-art box-supervised instance segmentation methods on COCO. The source code, pretrained models and datasets are available at https://github.com/hustvl/BoxCaseg. Xinggang Wang, Jiapei Feng, Bin Hu 0020, Longjin Ran, Xiaoxin Chen 0001, Wenyu Liu 0001 |
CVPR | 3 |
| 2021 | Learning generalizable deep feature using triplet-batch-center loss for person re-identification
Bin Hu 0020, Xinggang Wang |
Sci. China Inf. Sci. | 1 |