CUHK03-C is an evaluation set that consists of algorithmically generated corruptions applied to the CUHK03 test-set. These corruptions consist of Noise: Gaussian, shot,
impulse, and speckle; Blur: defocus, frosted glass, motion, zoom, and Gaussian; Weather: snow, frost, fog, brightness, spatter, and rain; Digital: contrast, elastic, pixel, JPEG compression, and saturate. Each corruption has five severity levels, resulting in 100 distinct corruptions.
Variants: CUHK03-C
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Person Re-Identification | CIL (ResNet-50) | Benchmarks for Corruption Invariant Person … | 2021-11-01 |
| Person Re-Identification | CaceNet | Devil's in the Details: Aligning … | 2020-09-11 |
| Person Re-Identification | MGN | Learning Diverse Features with Part-Level … | 2020-01-21 |
| Person Re-Identification | AGW (ResNet-50) | Deep Learning for Person Re-identification: … | 2020-01-13 |
| Person Re-Identification | RRID | Relation Network for Person Re-identification | 2019-11-21 |
| Person Re-Identification | MHN | Mixed High-Order Attention Network for … | 2019-08-16 |
| Person Re-Identification | Pyramid | Pyramidal Person Re-IDentification via Multi-Loss … | 2018-10-29 |
| Person Re-Identification | Aligned++ | AlignedReID: Surpassing Human-Level Performance in … | 2017-11-22 |
Recent papers with results on this dataset: