Saral Shiksha Yojna
Courses/Computer Vision

Computer Vision

CSE471
Prof. Makarand Tapaswi + Prof. Charu Sharma•Spring 2025-26•4 credits
Revision Notes/Unit 4 — 3D Data (PointNet, DGCNN, MeshCNN)

Unit 4 — 3D Data (PointNet, DGCNN, MeshCNN)

Why CNNs break on point clouds, voxelization's curse of dimensionality, PointNet's permutation-invariance theorem, hierarchical and graph-based variants, and operating on meshes directly via MeshCNN.

3D Representations — VoxNet, PointNet, PointNet++, DGCNN, MeshCNN

14 min

Point clouds break CNNs because CNNs assume a regular grid. The two big ideas are (1) voxelize and use 3D conv (memory-heavy, low resolution), or (2) build symmetric, permutation-invariant operators that act directly on the unordered point set.