KEYWORD |
Unsupervised/Semi-Supervised Video classification
Thesis in external company
keywords IA, ARTIFICIAL INTELLIGENCE, DEEP LEARNING
Reference persons BARTOLOMEO MONTRUCCHIO
External reference persons Enrico Busto enrico.busto@add-for.com
Research Groups GR-09 - GRAphics and INtelligent Systems - GRAINS
Thesis type RESEARCH
Description When there are hundreds or thousands of cameras producing video streams all day long it is very useful to have an algorithm that analyzes such streams instead of a human. Today such technology exists and is called convolutional neural networks for video classification [1]. The downside of such neural networks is that we have a fixed number of cases on which the net is trained which is ok for benchmarking our algorithm on a specific dataset but not for real life applications such as security cameras where we don't know specifically for which scene the algorithm should give an alert signal. So we need to produce an abstract representation of the video scene (embedding)[2] and to classify it in an unsupervised way [3].
[1] https://arxiv.org/pdf/1705.07750.pdf
[2] https://arxiv.org/pdf/1810.06951.pdf
[3] https://arxiv.org/pdf/1810.06951.pdf
[4] http://charuaggarwal.net/ICDE16_research_420.pdf
Required skills Python, advanced math, abstraction skills, exp. w. at least one among TensorFlow / PyTorch
Deadline 01/06/2020
PROPONI LA TUA CANDIDATURA