NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:3786
Title:GIFT: Learning Transformation-Invariant Dense Visual Descriptors via Group CNNs


		
This paper generated some disagreement among reviewers. The main strength is the novel technical contribution which can inspire others along similar directions. The main weakness is the lack of experimental evaluations that were suggested by the reviewers (additional datasets, baselines, and ablations). We hope that these can be included for the final revision.