English

PProCRC: Probabilistic Collaboration of Image Patches

Computer Vision and Pattern Recognition 2020-11-11 v3

Abstract

We present a conditional probabilistic framework for collaborative representation of image patches. It incorporates background compensation and outlier patch suppression into the main formulation itself, thus doing away with the need for pre-processing steps to handle the same. A closed form non-iterative solution of the cost function is derived. The proposed method (PProCRC) outperforms earlier CRC formulations: patch based (PCRC, GP-CRC) as well as the state-of-the-art probabilistic (ProCRC and EProCRC) on three fine-grained species recognition datasets (Oxford Flowers, Oxford-IIIT Pets and CUB Birds) using two CNN backbones (Vgg-19 and ResNet-50).

Keywords

Cite

@article{arxiv.1903.09123,
  title  = {PProCRC: Probabilistic Collaboration of Image Patches},
  author = {Tapabrata Chakraborti and Brendan McCane and Steven Mills and Umapada Pal},
  journal= {arXiv preprint arXiv:1903.09123},
  year   = {2020}
}
R2 v1 2026-06-23T08:15:22.154Z