珠宝
发表于 2025-3-23 13:38:59
http://reply.papertrans.cn/24/2343/234262/234262_11.png
glacial
发表于 2025-3-23 16:45:14
,Weakly Supervised Object Localization Through Inter-class Feature Similarity and Intra-class Appearused features for WSOL. However, existing CAM-based methods tend to excessively pursue discriminative features for object recognition and hence ignore the feature similarities among different categories, thereby leading to CAMs incomplete for object localization. In addition, CAMs are sensitive to b
倒转
发表于 2025-3-23 18:56:45
,Active Learning Strategies for Weakly-Supervised Object Detection,formance gap between them. We propose to narrow this gap by fine-tuning a base pre-trained weakly-supervised detector with a few fully-annotated samples automatically selected from the training set using “box-in-box” (BiB), a novel active learning strategy designed specifically to address the well-d
macabre
发表于 2025-3-23 23:44:01
http://reply.papertrans.cn/24/2343/234262/234262_14.png
独裁政府
发表于 2025-3-24 04:35:37
http://reply.papertrans.cn/24/2343/234262/234262_15.png
Aerophagia
发表于 2025-3-24 08:09:14
,Unsupervised Visual Representation Learning by Synchronous Momentum Grouping,asses the vanilla supervised learning. Two mainstream unsupervised learning schemes are the instance-level contrastive framework and clustering-based schemes. The former adopts the extremely fine-grained instance-level discrimination whose supervisory signal is not efficient due to the false negativ
使服水土
发表于 2025-3-24 12:09:34
Improving Few-Shot Part Segmentation Using Coarse Supervision,oit coarse labels such as figure-ground masks and keypoint locations that are readily available for some categories to improve part segmentation models. A key challenge is that these annotations were collected for different tasks and with different labeling styles and cannot be readily mapped to the
话
发表于 2025-3-24 17:00:13
,What to Hide from Your Students: Attention-Guided Masked Image Modeling,e token masking differs from token masking in text, due to the amount and correlation of tokens in an image. In particular, to generate a challenging pretext task for MIM, we advocate a shift from random masking to informed masking. We develop and exhibit this idea in the context of distillation-bas
昏迷状态
发表于 2025-3-24 22:36:56
Pointly-Supervised Panoptic Segmentation,evel labels used by fully supervised methods, point-level labels only provide a single point for each target as supervision, significantly reducing the annotation burden. We formulate the problem in an end-to-end framework by simultaneously generating panoptic pseudo-masks from point-level labels an
Microaneurysm
发表于 2025-3-25 02:30:22
http://reply.papertrans.cn/24/2343/234262/234262_20.png