5LSH0 Computer Vision AI and 3D Data Processing

Classical Computer Vision

1. Feature Extraction & Matching — Color spaces, edges, SIFT, HOG, Harris, Hough 2. Classification — Clustering — K-Means, GMM, EM, PCA 3. Classification — Supervised — kNN, SVM, AdaBoost, Haar cascades

Deep Learning

4. Deep Learning Fundamentals — Neurons, backprop, gradient descent, loss functions 5. Classification with Deep Learning — CNNs, pooling, batch norm, dropout, transfer learning 5.1. CNN Architecture Landscape — AlexNet, VGG, Inception, ResNet, DenseNet 6. Object Detection — R-CNN, YOLO, SSD, FPN, RetinaNet

3D Geometry & Sensing

7. Camera Model — Pinhole, intrinsics/extrinsics, calibration, homography 8. 3D Sensors & Depth Sensing — Stereo, ToF, structured light, LiDAR, radar

3D Reconstruction

9A. Neural Radiance Fields (NeRF) — Volume rendering, ray marching, positional encoding 9B. 3D Gaussian Splatting — Gaussian primitives, splatting, spherical harmonics 10. SLAM — Epipolar geometry, ICP, TSDF, KinectFusion

3D Processing & Analysis

10. Autoencoders — Encoder-decoder, bottleneck, denoising AE 11. 3D Data Fusion — Multi-modal fusion, ICP registration, TSDF 12. 3D Data Analysis — PointNet, MVCNN, VoxNet, 3D segmentation

Video & Behaviour

13. Behaviour Analysis — Action recognition, anomaly detection, I3D, SlowFast