activity
20172024
most citedRTFormer: Efficient Design for Real-Time Semantic Segmentation with Transformer

76 citations · 400 across the 33 of their papers we have counts for

collaborators
Showing 2023Show all

10 papers · 1 filter

cs.CV20235 cited

HAP: Structure-Aware Masked Image Modeling for Human-Centric Perception

Junkun Yuan, Xinyu Zhang, Hao Zhou +12

Model pre-training is essential in human-centric perception. In this paper, we first introduce masked image modeling (MIM) as a pre-training approach for this task. Upon revisiting…

cs.CV2023

Accelerating Vision Transformers Based on Heterogeneous Attention Patterns

Deli Yu, Teng Xi, Jianwei Li +7

Recently, Vision Transformers (ViTs) have attracted a lot of attention in the field of computer vision. Generally, the powerful representative capacity of ViTs mainly benefits from…

cs.CV20233 cited

Group Pose: A Simple Baseline for End-to-End Multi-person Pose Estimation

Huan Liu, Qiang Chen, Zichang Tan +9

In this paper, we study the problem of end-to-end multi-person pose estimation. State-of-the-art solutions adopt the DETR-like framework, and mainly develop the complex decoder, e.…

cs.CV20231 cited

MataDoc: Margin and Text Aware Document Dewarping for Arbitrary Boundary

Beiya Dai, Xing li, Qunyi Xie +5

Document dewarping from a distorted camera-captured image is of great value for OCR and document understanding. The document boundary plays an important role which is more evident…

cs.CV20231 cited

Semi-DETR: Semi-Supervised Object Detection with Detection Transformers

Jiacheng Zhang, Xiangru Lin, Wei Zhang +6

We analyze the DETR-based framework on semi-supervised object detection (SSOD) and observe that (1) the one-to-one assignment strategy generates incorrect matching when the pseudo…

cs.CV20238 cited

Learning Structure-Guided Diffusion Model for 2D Human Pose Estimation

Zhongwei Qiu, Qiansheng Yang, Jian Wang +7

One of the mainstream schemes for 2D human pose estimation (HPE) is learning keypoints heatmaps by a neural network. Existing methods typically improve the quality of heatmaps by c…