Ongoing reads - Interesting Sections - Applied ML - Vision

From this post onwards will review every section of Applied ML and capture the important reads.

Interesting Read#1 - RepNet: Counting Repetitions in Videos

For each frame, we then use Transformers to predict the period of repetition and the periodicity

• Repnet can be used to find number of product interactions
• Measure numbers of counts of physical activity
• Analyze count echocardiogram videos 

Interesting Read - 2 On-device Supermarket Product Recognition

• OCR + Text + N-Grams + Jaccard similarity coefficient

Interesting Read #3 - Drone control via gestures using MediaPipe Hands

Interesting Read #4 - Categorizing Listing Photos at Airbnb

• Dynamically cluster image captions instead of using rule-based heuristics
• One important note is that we built in production multiple binary-class models for different room types instead of building a multi-class model to cover all room types.

Keep Exploring!!!