PedFormer: Pedestrian Behavior Prediction via Cross-Modal Attention Modulation and Gated Multitask Learning.
Amir RasouliIuliia KotserubaPublished in: CoRR (2022)
Keyphrases
- cross modal
- multitask learning
- multi modal
- gaussian processes
- multi task
- learning tasks
- multimedia retrieval
- multi task learning
- multimedia databases
- machine learning
- visual recognition
- visual data
- labeled data
- semi supervised
- object detection
- image retrieval
- learned models
- information retrieval
- pedestrian detection
- human pose
- image sequences
- data sets