Výsledky vyhledávání

Report

Efficient Non-Exemplar Class-Incremental Learning with Retrospective Feature Synthesis

Autor: Bai, Liang, Song, Hong, Lin, Yucong, Fu, Tianyu, Xiao, Deqiang, Ai, Danni, Fan, Jingfan, Yang, Jian

Despite the outstanding performance in many individual tasks, deep neural networks suffer from catastrophic forgetting when learning from continuous data streams in real-world scenarios. Current Non-Exemplar Class-Incremental Learning (NECIL) methods

Externí odkaz: http://arxiv.org/abs/2411.01465

Zobrazit plný text záznamu

Report

Double-Shot 3D Shape Measurement with a Dual-Branch Network

Autor: Lei, Mingyang, Fan, Jingfan, Shao, Long, Song, Hong, Xiao, Deqiang, Ai, Danni, Fu, Tianyu, Gu, Ying, Yang, Jian

The structured light (SL)-based 3D measurement techniques with deep learning have been widely studied, among which speckle projection profilometry (SPP) and fringe projection profilometry (FPP) are two popular methods. However, they generally use a s

Externí odkaz: http://arxiv.org/abs/2407.14198

Zobrazit plný text záznamu

Report

High-Resolution Boundary Detection for Medical Image Segmentation with Piece-Wise Two-Sample T-Test Augmented Loss

Autor: Lin, Yucong, Su, Jinhua, Li, Yuhang, Wei, Yuhao, Yan, Hanchao, Zhang, Saining, Luo, Jiaan, Ai, Danni, Song, Hong, Fan, Jingfan, Fu, Tianyu, Xiao, Deqiang, Wang, Feifei, Hou, Jue, Yang, Jian

Deep learning methods have contributed substantially to the rapid advancement of medical image segmentation, the quality of which relies on the suitable design of loss functions. Popular loss functions, including the cross-entropy and dice losses, of

Externí odkaz: http://arxiv.org/abs/2211.02419

Zobrazit plný text záznamu