AdvMono3D: Advanced Monocular 3D Object Detection with Depth-Aware Robust Adversarial Training.
Xingyuan LiJinyuan LiuLong MaXin FanRisheng LiuPublished in: CoRR (2023)
Keyphrases
- d objects
- pose estimation
- viewpoint
- multi view
- object recognition
- partial occlusion
- depth cues
- multiple views
- three dimensional
- range data
- line drawings
- depth images
- object classes
- shape descriptors
- d mesh
- physically plausible
- depth information
- object detection
- depth map
- human body
- geometric models
- stereo images
- training examples
- three dimensional objects
- object features
- image contours
- machine learning
- object views
- pose normalization
- hough transform