BEV-CLIP: Multi-modal BEV Retrieval Methodology for Complex Scene in Autonomous Driving.
Dafeng WeiTian GaoZhengyu JiaChangwei CaiChengkai HouPeng JiaFu LiuKun ZhanJingchen FanYixing ZhaoYang WangPublished in: CoRR (2024)
Keyphrases
- multi modal
- complex scenes
- autonomous driving
- cross modal
- video search
- grand challenge
- multi modality
- stereo vision
- audio visual
- semantic concepts
- high dimensional
- image annotation
- multimedia databases
- uni modal
- multimedia retrieval
- multiple objects
- computer graphics
- image retrieval
- retrieval systems
- relevance feedback
- urban traffic
- image registration