Improving Robotic Grasping on Monocular Images Via Multi-Task Learning and Positional Loss.
William PrewToby P. BreckonMagnus BordewichUlrik R. BeierholmPublished in: ICPR (2020)
Keyphrases
- em algorithm
- multi task learning
- monocular images
- multi task
- maximum likelihood
- human body
- pose estimation
- learning tasks
- gaussian processes
- learning problems
- generative model
- transfer learning
- learning algorithm
- high order
- depth estimation
- theoretical analysis
- vision system
- human pose
- computer vision
- markov random field
- human motion
- temporal sequences