Vision-enhanced Peg-in-Hole for automotive body parts using semantic image segmentation and object detection.
Monica SileoNicola CapeceMonica GruossoMichelangelo NigroDomenico Daniele BloisiFrancesco PierriUgo ErraPublished in: Eng. Appl. Artif. Intell. (2024)
Keyphrases
- body parts
- object detection
- computer vision
- image segmentation
- human body
- pose estimation
- action recognition
- pedestrian detection
- human motion
- object segmentation
- human pose estimation
- motion tracking
- video sequences
- image processing
- human pose
- face detection
- scene recognition
- video frames
- object categories
- human detection
- markov random field
- vision system
- human body parts
- spatial and temporal
- upper body
- human posture
- volumetric reconstruction
- active contours
- graph cuts
- multi class
- spatio temporal
- human actions
- shape prior
- degrees of freedom
- background subtraction
- segmentation method
- object localization
- level set
- body posture
- image sequences
- three dimensional
- metadata