Crosswalk Recognition Through Point-Cloud Processing and Deep-Learning Suited to a Wearable Mobility Aid for the Visually Impaired.
Matteo PoggiLuca NanniStefano MattocciaPublished in: ICIAP Workshops (2015)
Keyphrases
- point cloud
- deep learning
- unsupervised learning
- structure from motion
- unsupervised feature learning
- pattern recognition
- point sets
- gesture recognition
- object recognition
- point cloud data
- machine learning
- laser scanner
- weakly supervised
- mental models
- appearance model
- single image
- human computer interaction
- stereo camera
- expectation maximization
- supervised learning
- feature extraction
- image sequences