**Overview** I've been playing around with computer vision for the past 18 months and tried different approaches to solve a "simple" problem: detect the content of my fridge. In short : I'd like to duplicated this project https://github.com/stphnhng/SmartFridgeImageRecognition After following a bunch of tutorials and Youtube vids, I never got to a point where the model was functional. **Challenge** The dataset that was uploaded consist of 2 zipped folders - containing 2 different products. The products are separated in the folders. The challenge would be to build a model that will be able to detect these objects - with moderate accuracy - in a webcam live chat, with about 20 fps. Thus, picking up the cola can or the chocolate bar is in the picture - without the major impact on the video picture. The initial plan is just to run the model on my laptop's webcam - and later move it to a Raspberry Pi3 depending on feasibility. I'm sure there are guys out there that do this type of work in their sleep - but this project has been haunting me. **Involvement** I'd be happy to assist where possible - specifically with labelling as I know this is extremely time consuming - would need some guidance from you thought. As this is my first dataset uploaded - the requirements for some traffic is still in a testing phase, and I'll be happy to get some guidance on what can make this project more "eye catching" **Acknowledgements** Some references used: http://www.maths.lth.se/vision/publdb/reports/pdf/farnstrom-johansson-etal-sia-01.pdf https://blogs.technet.microsoft.com/machinelearning/2016/10/25/how-to-train-a-deep-learned-object-detection-model-in-cntk/ https://github.com/stphnhng/SmartFridgeImageRecognition We wouldn't be here without the help of others. If you owe any attributions or thanks, include them here along with any citations of past research. **Inspiration** This is a super exciting field and I'm learning a lot - but getting stuck on a simple program makes for extreme frustration and time wasting on different "off the shelf solutions" by major players ("G" "M" and "A").