Object detection (with ROIs) in Mocha.jl?



I’m trying to get bounding-box style object detection running through Mocha.jl and wondering if anyone is aware of any work on this front? Mocha has tidy image classification examples, but I haven’t found an object detection example.

Knowing that Mocha.jl is based on Caffe I’ve found a lot of work implementing object detection in caffe, but the code seems to have diverged to the point that it’s not obvious how to implement these into Mocha.

Also, the state of the art in Caffe seems rather impressive, based on this agglomeration of packages that facebook have developed.

I’d love to find Mocha object detection work to build upon, given there much smarter people than me working on all this and my knowledge of deep learning isn’t particularly deep… Otherwise my plan is to ccall or if needed pycall into a package with a working example.