Congeal
发表于 2025-3-28 16:31:30
,RealPatch: A Statistical Matching Framework for Model Patching with Real Samples,n generative models for the target task. We demonstrate the effectiveness of RealPatch on three benchmark datasets, CelebA, Waterbirds and a subset of iWildCam, showing improvements in worst-case subgroup performance and in subgroup performance gap in binary classification. Furthermore, we conduct e
镶嵌细工
发表于 2025-3-28 20:41:12
,Background-Insensitive Scene Text Recognition with Text Semantic Segmentation, features. Additionally, we propose an efficient pipeline that utilizes Synthetic Initialization (SI) for STR models trained only on real data (1.7% of STR training data), instead of on both synthetic and real data from scratch. Experiments show that the proposed method can recognize text from compl
围巾
发表于 2025-3-29 01:49:36
http://reply.papertrans.cn/24/2343/234273/234273_43.png
nullify
发表于 2025-3-29 05:45:33
http://reply.papertrans.cn/24/2343/234273/234273_44.png
FAWN
发表于 2025-3-29 11:13:15
,TDAM: Top-Down Attention Module for Contextually Guided Feature Selection in CNNs, TDAM enhances the performance of CNNs across multiple object-recognition benchmarks and outperforms prominent attention modules while being more parameter and memory efficient. Further, TDAM-based models learn to “shift attention” by localizing individual objects or features at each computation ste
circumvent
发表于 2025-3-29 15:07:09
http://reply.papertrans.cn/24/2343/234273/234273_46.png
moratorium
发表于 2025-3-29 17:38:11
http://reply.papertrans.cn/24/2343/234273/234273_47.png
covert
发表于 2025-3-29 23:23:48
http://reply.papertrans.cn/24/2343/234273/234273_48.png
Itinerant
发表于 2025-3-30 03:40:16
http://reply.papertrans.cn/24/2343/234273/234273_49.png
spinal-stenosis
发表于 2025-3-30 06:34:35
http://reply.papertrans.cn/24/2343/234273/234273_50.png