Strike (with) a Pose: Neural Networks Are Easily Fooled by Strange Poses of Familiar Objects.
Michael A. AlcornQi LiZhitao GongChengfei WangLong MaiWei-Shinn KuAnh NguyenPublished in: CoRR (2018)
Keyphrases
- neural network
- d objects
- pose invariant face recognition
- human pose
- rigid objects
- pose variations
- pose estimation
- object geometry
- artificial neural networks
- partial occlusion
- genetic algorithm
- fuzzy logic
- multilayer perceptron
- multiple objects
- moving objects
- face recognition
- self organizing maps
- feed forward
- viewpoint
- body parts
- pattern recognition
- object representation
- image sequences
- three dimensional