Major project Automotive, EV & Aerospace Case Studio

Road object detection and tracking

Inspired by publicly reported work at Waymo

Detect and track cars, people, bikes and autos in driving video, and test how well it works on Indian roads.

The brief

Waymo has released one of the largest open datasets for self-driving research and publishes its safety research. Build a perception module: train an object detector on public driving data, add tracking across frames, and test how a model trained on Western roads performs on Indian roads with autos, two-wheelers and animals. Good looks like: mAP per class, tracking metrics, a clear drop analysis from Western to Indian data, and a demo video with a written safety case.

Read the case first

What Waymo has said in public about this kind of work, on its own pages:

Make it your own

Pick a narrower job than full self-driving: a driver-alert camera for buses, a pedestrian warning at a school gate, or counting vehicles at a junction. Decide which mistakes are dangerous for that job and judge the model on those.

AITeky has no link with Waymo, which has not reviewed or endorsed this brief. The brief only draws on what the company has published, uses open data rather than its data, and is a learning exercise: don't present your project as Waymo's work or system.

What you will build

Train a detector (YOLO or similar) on public driving data
Evaluate mAP per class
Add multi-object tracking across frames
Test on Indian road data and analyse the drop
Fine-tune on Indian data and measure the gain
Make a demo video and a short safety case

Free, open data to build it with

It takes a minute with your email. AiTeky Pass (₹999/month or ₹9,999/year) is free for your first 30 days.