1. The solution shall be evaluated on the architecture design of the entire application.
2. The solution shall be evaluated on the data pre-processing and inferences drawn.
3. The solution shall be evaluated based on the optimizations applied and the model accuracy on an unseen dataset.
4. The solution shall be evaluated based on testing with the realtime dataset.
5. The solution shall be evaluated based on the code readability, formatting and proper commenting whenever and wherever required.
6. The solution shall be evaluated based on the interactiveness and ease of use of the user interface and its aesthetics.
The solutions need to be submitted on or before the last date for submission. The solution submission should include a working application satisfying all the requirements along with full documentation and the video recording for the project as mentioned in the solution requirements section. The links for all code, documentations and videos can be provided below separately for each of the problem statements.
Please submit the code, documentation and video link one by one for each of the projects.
Individual problem statements can be accessed by opening the problem tab. For getting the detailed problem statement, attribute information and data, please click on the respective problem statement button to download the file.
To build a robust pedestrian detection system that can detect and count the people in real-time.
To build a robust Automatic Number plate recognition system that can detect moving vehicles in real-time traffic and extract the number plate(license plate) of the vehicles.
To build a robust face recognition system that can detect and track the face of the user.
To build a robust pose and activity recognizer system that can detect the poses and activity of the persons in real-time live streaming.