Pedestrian Trajectory Prediction in Heterogeneous Traffic Using Pose Keypoints-Based Convolutional Encoder-Decoder Network.
Kai ChenXiao SongXiaoxiang RenPublished in: IEEE Trans. Circuits Syst. Video Technol. (2021)
Keyphrases
- keypoints
- network traffic
- pose variations
- low complexity
- scale invariant
- decoding process
- video codec
- keypoint detection
- distributed video coding
- vehicle trajectory
- sift features
- multiscale
- image matching
- object detection
- feature detection
- feature extraction
- d objects
- error control
- sift descriptors
- visual words
- scale invariant feature transform
- bit rate
- wyner ziv
- mode decision
- road network
- noisy channel
- invariant features
- sparse coding