Výsledky vyhledávání

Report

Organizing Background to Explore Latent Classes for Incremental Few-shot Semantic Segmentation

Autor: Shan, Lianlei, Zhou, Wenzhang, Li, Wei, Ding, Xingyu

The goal of incremental Few-shot Semantic Segmentation (iFSS) is to extend pre-trained segmentation models to new classes via few annotated images without access to old training data. During incrementally learning novel classes, the data distribution

Externí odkaz: http://arxiv.org/abs/2405.19568

Zobrazit plný text záznamu

Report

Lifelong Learning and Selective Forgetting via Contrastive Strategy

Autor: Shan, Lianlei, Zhou, Wenzhang, Li, Wei, Ding, Xingyu

Lifelong learning aims to train a model with good performance for new tasks while retaining the capacity of previous tasks. However, some practical scenarios require the system to forget undesirable knowledge due to privacy issues, which is called se

Externí odkaz: http://arxiv.org/abs/2405.18663

Zobrazit plný text záznamu

Report

The Binary Quantized Neural Network for Dense Prediction via Specially Designed Upsampling and Attention

Autor: Ding, Xingyu, Shan, Lianlei, Zhao, Guiqin, Wu, Meiqi, Zhou, Wenzhang, Li, Wei

Deep learning-based information processing consumes long time and requires huge computing resources, especially for dense prediction tasks which require an output for each pixel, like semantic segmentation and salient object detection. There are main

Externí odkaz: http://arxiv.org/abs/2405.17776

Zobrazit plný text záznamu