ADfM-Net: An Adversarial Depth-From-Motion Network Based on Cross Attention and Motion Enhanced.
Yangqi LongHuimin YuChenfeng XuZhiqiang DengPublished in: IEEE Robotics Autom. Lett. (2023)
Keyphrases
- depth from motion
- visual sensor
- optical flow
- image sequences
- motion estimation
- motion model
- human motion
- visual odometry
- motion analysis
- motion field
- camera motion
- flow field
- video sequences
- motion segmentation
- motion parameters
- motion detection
- humanoid robot
- optic flow
- computer vision
- feature points
- three dimensional
- autonomous navigation
- ego motion
- velocity field
- multi frame
- moving camera
- dynamic scenes
- real time
- visual attention
- moving objects
- space time