VRAI (Vehicle Re-identification for Aerial Image)
Introduced by Wang et al. in Vehicle Re-identification in Aerial Imagery: Dataset and Approach
VRAI is a large-scale vehicle ReID dataset for UAV-based intelligent applications. The dataset consists of 137, 613 images of 13, 022 vehicle instances. The images of each vehicle instance are captured by cameras of two DJI consumer UAVs at different locations, with a variety of view angles and flight-altitudes (15m to 80m).
Source: Vehicle Re-identification in Aerial Imagery: Dataset and ApproachBenchmarks
| Trend | Task | Dataset Variant | Best Model | Paper | Code |
|---|---|---|---|---|---|
|
VRAI test-dev
|
Baseline Model
|
|||
|
VRAI test
|
Baseline Model
|
Papers
| Paper | Code | Results | Date | Stars |
|---|
