Learning Multi-Instance Enriched Image Representations via Non-Greedy Ratio Maximization of the l1-Norm Distances.

Kai Liu, Hua Wang, Feiping Nie, Hao Zhang

Browse the full CVPR paper archive.