Zobrazeno 1 - 10
of 17
pro vyhledávání: '"Lin, Xiangru"'
Autor:
Jiang, Yujiao, Liao, Qingmin, Wang, Zhaolong, Lin, Xiangru, Lu, Zongqing, Zhao, Yuxi, Wei, Hanqing, Ye, Jingrui, Zhang, Yu, Shao, Zhijing
Recovering photorealistic and drivable full-body avatars is crucial for numerous applications, including virtual reality, 3D games, and tele-presence. Most methods, whether reconstruction or generation, require large numbers of human motion sequences
Externí odkaz:
http://arxiv.org/abs/2405.19609
Autor:
Zhang, Jiacheng, Li, Jiaming, Lin, Xiangru, Zhang, Wei, Tan, Xiao, Han, Junyu, Ding, Errui, Wang, Jingdong, Li, Guanbin
We delve into pseudo-labeling for semi-supervised monocular 3D object detection (SSM3OD) and discover two primary issues: a misalignment between the prediction quality of 3D and 2D attributes and the tendency of depth supervision derived from pseudo-
Externí odkaz:
http://arxiv.org/abs/2403.17387
Autor:
Li, Jiaming, Lin, Xiangru, Zhang, Wei, Tan, Xiao, Li, Yingying, Han, Junyu, Ding, Errui, Wang, Jingdong, Li, Guanbin
Current semi-supervised object detection (SSOD) algorithms typically assume class balanced datasets (PASCAL VOC etc.) or slightly class imbalanced datasets (MS-COCO, etc). This assumption can be easily violated since real world datasets can be extrem
Externí odkaz:
http://arxiv.org/abs/2403.15127
Autor:
Shao, Zhijing, Wang, Zhaolong, Li, Zhuang, Wang, Duotun, Lin, Xiangru, Zhang, Yu, Fan, Mingming, Wang, Zeyu
We present SplattingAvatar, a hybrid 3D representation of photorealistic human avatars with Gaussian Splatting embedded on a triangle mesh, which renders over 300 FPS on a modern GPU and 30 FPS on a mobile device. We disentangle the motion and appear
Externí odkaz:
http://arxiv.org/abs/2403.05087
Autor:
Zhang, Jiacheng, Lin, Xiangru, Zhang, Wei, Wang, Kuo, Tan, Xiao, Han, Junyu, Ding, Errui, Wang, Jingdong, Li, Guanbin
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 ground-truth bounding box is inaccurate, leading to training inefficie
Externí odkaz:
http://arxiv.org/abs/2307.08095
Autor:
Liu, Chang, Zhang, Weiming, Lin, Xiangru, Zhang, Wei, Tan, Xiao, Han, Junyu, Li, Xiaomao, Ding, Errui, Wang, Jingdong
With basic Semi-Supervised Object Detection (SSOD) techniques, one-stage detectors generally obtain limited promotions compared with two-stage clusters. We experimentally find that the root lies in two kinds of ambiguities: (1) Selection ambiguity th
Externí odkaz:
http://arxiv.org/abs/2303.14960
Unsupervised domain adaptation (UDA) for semantic segmentation is a promising task freeing people from heavy annotation work. However, domain discrepancies in low-level image statistics and high-level contexts compromise the segmentation performance
Externí odkaz:
http://arxiv.org/abs/2301.01149
Unsupervised domain adaptation (UDA) in semantic segmentation is a fundamental yet promising task relieving the need for laborious annotation works. However, the domain shifts/discrepancies problem in this task compromise the final segmentation perfo
Externí odkaz:
http://arxiv.org/abs/2103.13041
Humans learn from life events to form intuitions towards the understanding of visual environments and languages. Envision that you are instructed by a high-level instruction, "Go to the bathroom in the master bedroom and replace the blue towel on the
Externí odkaz:
http://arxiv.org/abs/2103.12944
Given a training dataset composed of images and corresponding category labels, deep convolutional neural networks show a strong ability in mining discriminative parts for image classification. However, deep convolutional neural networks trained with
Externí odkaz:
http://arxiv.org/abs/1903.02827