Zobrazeno 1 - 10
of 105
pro vyhledávání: '"yang, Kaicheng"'
Autor:
Yang, Kaicheng, Gu, Tiancheng, An, Xiang, Jiang, Haiqiang, Dai, Xiangzi, Feng, Ziyong, Cai, Weidong, Deng, Jiankang
Contrastive Language-Image Pre-training (CLIP) has achieved excellent performance over a wide range of tasks. However, the effectiveness of CLIP heavily relies on a substantial corpus of pre-training data, resulting in notable consumption of computat
Externí odkaz:
http://arxiv.org/abs/2408.09441
Contrastive Language Image Pre-training (CLIP) has recently demonstrated success across various tasks due to superior feature representation empowered by image-text contrastive learning. However, the instance discrimination method used by CLIP can ha
Externí odkaz:
http://arxiv.org/abs/2407.17331
Autor:
Ran, Zimin, Ren, Xingyu, An, Xiang, Yang, Kaicheng, Dai, Xiangzi, Feng, Ziyong, Guo, Jia, Zhu, Linchao, Deng, Jiankang
Recent 3D face reconstruction methods have made significant progress in shape estimation, but high-fidelity facial albedo reconstruction remains challenging. Existing methods depend on expensive light-stage captured data to learn facial albedo maps.
Externí odkaz:
http://arxiv.org/abs/2406.13149
Autor:
Gu, Tiancheng, Yang, Kaicheng, An, Xiang, Feng, Ziyong, Liu, Dongnan, Cai, Weidong, Deng, Jiankang
Contrastive Language-Image Pre-training (CLIP) has significantly improved performance in various vision-language tasks by expanding the dataset with image-text pairs obtained from websites. This paper further explores CLIP from the perspectives of da
Externí odkaz:
http://arxiv.org/abs/2406.06973
Autor:
Wu, Yuhan, Wu, Hanbo, Liu, Xilai, Zhao, Yikai, Yang, Tong, Yang, Kaicheng, Wang, Sha, Miao, Lihua, Xie, Gaogang
To approximate sums of values in key-value data streams, sketches are widely used in databases and networking systems. They offer high-confidence approximations for any given key while ensuring low time and space overhead. While existing sketches are
Externí odkaz:
http://arxiv.org/abs/2406.00376
This paper presents the winning solution for the 1st SkatingVerse Challenge. We propose a method that involves several steps. To begin, we leverage the DINO framework to extract the Region of Interest (ROI) and perform precise cropping of the raw vid
Externí odkaz:
http://arxiv.org/abs/2404.14032
Medical visual question answering (Med-VQA) aims to automate the prediction of correct answers for medical images and questions, thereby assisting physicians in reducing repetitive tasks and alleviating their workload. Existing approaches primarily f
Externí odkaz:
http://arxiv.org/abs/2404.13039
Autor:
Yang, Kaicheng, Deng, Jiankang, An, Xiang, Li, Jiawei, Feng, Ziyong, Guo, Jia, Yang, Jing, Liu, Tongliang
Contrastive Language-Image Pre-training (CLIP) has significantly boosted the performance of various vision-language tasks by scaling up the dataset with image-text pairs collected from the web. However, the presence of intrinsic noise and unmatched i
Externí odkaz:
http://arxiv.org/abs/2308.08428
Publikováno v:
口腔疾病防治, Vol 32, Iss 10, Pp 765-771 (2024)
Objective To investigate the expression of trophoblast cell-surface antigen 2 (TROP2) in salivary adenoid cystic carcinoma (SACC) in order to analyze its relationship with TROP2 expression and clinicopathological features, as well as to clarify the c
Externí odkaz:
https://doaj.org/article/6b04dfecc81348ee824a6ab3b7138074
Autor:
An, Xiang, Deng, Jiankang, Yang, Kaicheng, Li, Jaiwei, Feng, Ziyong, Guo, Jia, Yang, Jing, Liu, Tongliang
Modern image retrieval methods typically rely on fine-tuning pre-trained encoders to extract image-level descriptors. However, the most widely used models are pre-trained on ImageNet-1K with limited classes. The pre-trained feature representation is
Externí odkaz:
http://arxiv.org/abs/2304.05884