1 code implementation • 12 Mar 2024 • Lei Zhu, Fangyun Wei, Yanye Lu
To achieve this, we present the Vision-to-Language Tokenizer, abbreviated as V2T Tokenizer, which transforms an image into a ``foreign language'' with the combined aid of an encoder-decoder, the LLM vocabulary, and a CLIP model.
1 code implementation • 27 Feb 2024 • Xinliang Zhang, Lei Zhu, Hangzhou He, Lujia Jin, Yanye Lu
In this study, we propose a class-driven scribble promotion network, which utilizes both scribble annotations and pseudo-labels informed by image-level classes and global semantics for supervision.
no code implementations • 21 Sep 2023 • Shuang Zeng, Lei Zhu, Xinliang Zhang, Zifeng Tian, Qian Chen, Lujia Jin, Jiayi Wang, Yanye Lu
In this work, we propose a novel asymmetric contrastive learning framework named JCL for medical image segmentation with self-supervised pre-training.
no code implementations • 9 Aug 2023 • Lei Zhu, Hangzhou He, Xinliang Zhang, Qian Chen, Shuang Zeng, Qiushi Ren, Yanye Lu
Existing methods adopt an online-trained classification branch to provide pseudo annotations for supervising the segmentation branch.
no code implementations • 18 Feb 2023 • Lujia Jin, Shi Zhao, Lei Zhu, Qian Chen, Yanye Lu
Therefore, it is necessary to avoid the restriction of clean labels and make full use of noisy data for model training.
1 code implementation • 16 Jul 2022 • Lei Zhu, Qian Chen, Lujia Jin, Yunfei You, Yanye Lu
Classification activation map (CAM), utilizing the classification structure to generate pixel-wise localization maps, is a crucial mechanism for weakly supervised object localization (WSOL).
1 code implementation • CVPR 2022 • Lei Zhu, Qi She, Qian Chen, Yunfei You, Boyu Wang, Yanye Lu
To avoid this problem, this work provides a novel perspective that models WSOL as a domain adaption (DA) task, where the score estimator trained on the source/image domain is tested on the target/pixel domain to locate objects.
2 code implementations • IEEE Transactions on Medical Imaging 2022 • Mufeng Geng, Xiangxi Meng, Jiangyuan Yu, Lei Zhu, Lujia Jin, Zhe Jiang, Bin Qiu, Hui Li, Hanjing Kong, Jianmin Yuan, Kun Yang, Hongming Shan, Hongbin Han, Zhi Yang, Qiushi Ren, Yanye Lu
In this study, we propose a simple yet effective strategy, the content-noise complementary learning (CNCL) strategy, in which two deep learning predictors are used to learn the respective content and noise of the image dataset complementarily.
1 code implementation • 29 Dec 2021 • Lei Zhu, Qi She, Qian Chen, Xiangxi Meng, Mufeng Geng, Lujia Jin, Zhe Jiang, Bin Qiu, Yunfei You, Yibao Zhang, Qiushi Ren, Yanye Lu
In our B-CAM, two image-level features, aggregated by pixel-level features of potential background and object locations, are used to purify the object feature from the object-related background and to represent the feature of the pure-background sample, respectively.
1 code implementation • ICCV 2021 • Lei Zhu, Qi She, Duo Li, Yanye Lu, Xuejing Kang, Jie Hu, Changhu Wang
The nonlocal-based blocks are designed for capturing long-range spatial-temporal dependencies in computer vision tasks.
1 code implementation • 23 Jun 2021 • Mengdi Gao, Ximeng Feng, Mufeng Geng, Zhe Jiang, Lei Zhu, Xiangxi Meng, Chuanqing Zhou, Qiushi Ren, Yanye Lu
BLRM utilizes maximum a posteriori probability (MAP) in the Bayesian statistics and the exponentially time-weighted technique to selectively correct the labels of noisy images.
1 code implementation • CVPR 2021 • Lei Zhu, Qi She, Bin Zhang, Yanye Lu, Zhilin Lu, Duo Li, Jie Hu
Superpixel is generated by automatically clustering pixels in an image into hundreds of compact partitions, which is widely used to perceive the object contours for its excellent contour adherence.