Experimenting with Multi-views CNNs with Caffe, work under process.

Ideally, we want to demonstrate that the performance of CNNs for image classification can be improved by providing multiple perspectives --of the same frame-- to the network. Otherwise, one can reduce the depth (and thus the computations) of a CNN and keep tolerable classification accuracy by providing multi-perspective inputs.

• Original paper from Hang et al.
• Original git repository
• Dataset : ModelNet40v1
• Network : Alexnet topology for now, will be experimenting lighter CNNs after ...

1. Evaluate accuracy of vanilla alexnet on modelnetv1, 1.5% is clearly bad
2. Study with 3:12 number of views
3. Explore/define where to put the view-pooling layer

1. Shitty ModelNet40v1 Dataset: CAD Images, not sure it with work in real world Images
2. Multi-view images of large objects in demos