Zobrazeno 1 - 10
of 42
pro vyhledávání: '"Chang, Xiaobin"'
Autor:
Yao, Zhongren, Chang, Xiaobin
Exemplar-free class-incremental learning (EFCIL) presents a significant challenge as the old class samples are absent for new task learning. Due to the severe imbalance between old and new class samples, the learned classifiers can be easily biased t
Externí odkaz:
http://arxiv.org/abs/2409.13275
Autor:
Liu, Feng, Chang, Xiaobin
Semantic image synthesis aims to generate high-quality images given semantic conditions, i.e. segmentation masks and style reference images. Existing methods widely adopt generative adversarial networks (GANs). GANs take all conditional inputs and di
Externí odkaz:
http://arxiv.org/abs/2403.13378
Continual learning empowers models to adapt autonomously to the ever-changing environment or data streams without forgetting old knowledge. Prompt-based approaches are built on frozen pre-trained models to learn the task-specific prompts and classifi
Externí odkaz:
http://arxiv.org/abs/2403.08568
Deep neural networks often severely forget previously learned knowledge when learning new knowledge. Various continual learning (CL) methods have been proposed to handle such a catastrophic forgetting issue from different perspectives and achieved su
Externí odkaz:
http://arxiv.org/abs/2402.18086
Autor:
Chen, Xiuwei, Chang, Xiaobin
Class incremental learning (CIL) aims to recognize both the old and new classes along the increment tasks. Deep neural networks in CIL suffer from catastrophic forgetting and some approaches rely on saving exemplars from previous tasks, known as the
Externí odkaz:
http://arxiv.org/abs/2308.15236
Autor:
Chen, Xiuwei, Chang, Xiaobin
The rehearsal strategy is widely used to alleviate the catastrophic forgetting problem in class incremental learning (CIL) by preserving limited exemplars from previous tasks. With imbalanced sample numbers between old and new classes, the classifier
Externí odkaz:
http://arxiv.org/abs/2308.13305
Camera-Conditioned Stable Feature Generation for Isolated Camera Supervised Person Re-IDentification
To learn camera-view invariant features for person Re-IDentification (Re-ID), the cross-camera image pairs of each person play an important role. However, such cross-view training samples could be unavailable under the ISolated Camera Supervised (ISC
Externí odkaz:
http://arxiv.org/abs/2203.15210
Autor:
Qiu, Yiqiao, Shen, Yixing, Sun, Zhuohao, Zheng, Yanchong, Chang, Xiaobin, Zheng, Weishi, Wang, Ruixuan
Publikováno v:
Pattern Recognition (2023) 109383
Continually learning to segment more and more types of image regions is a desired capability for many intelligent systems. However, such continual semantic segmentation suffers from the same catastrophic forgetting issue as in continual classificatio
Externí odkaz:
http://arxiv.org/abs/2203.07667
Dynamic Time Warping (DTW) is widely used for temporal data processing. However, existing methods can neither learn the discriminative prototypes of different classes nor exploit such prototypes for further analysis. We propose Discriminative Prototy
Externí odkaz:
http://arxiv.org/abs/2103.09458
Autor:
Qiu, Yiqiao, Shen, Yixing, Sun, Zhuohao, Zheng, Yanchong, Chang, Xiaobin, Zheng, Weishi, Wang, Ruixuan
Publikováno v:
In Pattern Recognition June 2023 138
