ujjawalcse / Retail-Visual-Analysis

Analize and Provide Useful Insight for Retail Store using Camera Feed

Retail-Visual-Analysis

Analize and Provide Useful Insight for Retail Store using Camera Feed

1. Timely footfall detection

• Used quit fast,accurate,light,robust and efficient deep learning model trained on people detection in a video feed.

2. Demography detection

• Used FairFace deep learning model to detect Age,Gender,Race etc given a facial portion of a video frame.

3. Hot-zone detection

• Customer engagement in a specific zone of the retail store is evaluated.

4. Repeating customer detection

• Used Face Recognition deep learning model to recognize repeated customer if any.

5. Shelf Analysis

• Annotated and Prepared around 8k shelf packet image data
• Trained YoloV5 model to detect packets in a shelf, It leads to the number of packet in the shelf
• Trained YoloV5 model to detect empty spaces in a shelf,It leads to the number of empty spaces in the shelf

6. Customer activity recognition near the shelf

• Used temporal deep learning model (3D convolution, LSTM combined with Convolution etc) to recognize differect sets of activity perfomed nearby the shelf