Choice 1: Density Map

In this project, you will build a convolutional neural network that estimates the number of individuals in a crowded scene by analyzing the spatial density of people in an image. Your implementation will follow the approach described in the paper “CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes” by Li et al. (2018). CSRNet extends a pretrained VGG16 backbone (frontend) with dilated convolutional layers (backend), enabling the model to capture larger contextual regions without losing spatial resolution. The model is trained to generate a density map, where the integral over its values corresponds to the estimated count of individuals in the input image.

Difficulty	Suggested Tutorials
Easy	Convolutional Networks, Transfer Learning

Requirements: A GPU with at least 8GB of memory is recommended. Training on a CPU is possible but will be significantly slower.

---

Grading

The project will be graded based on the following criteria. Points for each activity are awarded based on quality and completeness (partial credit possible).

Activity	Points (max)
Build a preprocessing pipeline for the ShangaiTech dataset	4
Implement the CSRNet architecture on top of VGG16	4
Train the backend while keeping the frontend frozen	2
Fine-tune the entire network end-to-end	3
Evaluate model performance using MAE	4
Presentation (clarity & demo)	3
Total	0-20