Course tree
Open the book you want to read, then explore its chapters, sub-chapters, notes and paper guides. Books and chapters start collapsed.
Foundations
2 main chapters
Image Processing
13 main chapters
- IP 01 · Introduction
2 optional pages: Supplementary notes 2
- IP 03 · Intensity Transformations and Spatial Filtering
8 optional pages: Supplementary notes 5 · Paper guides 3
- IP 04 · Filtering in the Frequency Domain
3 optional pages: Supplementary notes 2 · Paper guides 1
- IP 05 · Image Restoration and Reconstruction
10 optional pages: Supplementary notes 7 · Paper guides 3
- IP 07 · Wavelet and Other Image Transforms
5 optional pages: Supplementary notes 4 · Paper guides 1
- IP 08 · Image Compression and Watermarking
4 optional pages: Supplementary notes 3 · Paper guides 1
- IP 09 · Morphological Image Processing
10 optional pages: Supplementary notes 9 · Paper guides 1
- IP 11 · Image Segmentation II: Active Contours — Snakes and Level Sets
11 optional pages: Supplementary notes 9 · Paper guides 2
- IP 12 · Feature Extraction
12 optional pages: Supplementary notes 10 · Paper guides 2
Supplementary notes- The Moore boundary-following procedure
- Slope chain codes
- Minimum-perimeter polygons in more detail
- The medial axis transform
- Computing shape numbers of a given order
- Moments of a 1-D boundary function
- Euler’s formula for polygonal networks
- Keypoint localization
- RANSAC: fitting a model despite outliers
- Learned features in more detail
Natural Language Processing
12 main chapters
- NLP 02 · Linguistic Essentials
1 optional page: Paper guides 1
- NLP 03 · Neural Networks for NLP
1 optional page: Paper guides 1
- NLP 05 · Recurrent Neural Networks for Language
1 optional page: Paper guides 1
- NLP 07 · Attention Mechanism and Transformers
1 optional page: Paper guides 1
- NLP 12 · Syntactic and Semantic Tasks
1 optional page: Paper guides 1
Pattern Recognition
10 main chapters
- PR 06 · Template Matching and Structured Recognition
1 optional page: Supplementary notes 1
- PR 09 · Ensembles, Validation and Cluster Validity
1 optional page: Paper guides 1
Machine Learning
11 main chapters
ML 01 · Introduction to Machine Learning3 sub-chapters
1 optional page: Paper guides 1
- ML 01A · Linear Algebra for Learning
1 optional page: Paper guides 1
- ML 01B · Probability and Information Theory
1 optional page: Paper guides 1
- ML 01C · Numerical Computation and Gradient Checks
1 optional page: Paper guides 1
- ML 02 · Neural Networks from Perceptron to Deep Networks
1 optional page: Paper guides 1
- ML 03 · Convolution, Image Filtering and CNNs
1 optional page: Paper guides 1
- ML 04 · Recurrent and Recursive Networks
1 optional page: Paper guides 1
- ML 05 · Attention and Vision Transformers
1 optional page: Paper guides 1
ML 06 · Representation Learning and Self-Supervision1 sub-chapter
1 optional page: Paper guides 1
- ML 06A · Linear Factor Models and Autoencoders
1 optional page: Paper guides 1
- ML 08 · Diffusion Models and Flow Matching
6 optional pages: Paper guides 6
Paper guides- DDPM — noise prediction and the quality–likelihood tradeoff
- Flow Matching — regress a velocity, then integrate it
- MeanFlow — average velocity over an interval
- Improved MeanFlow — prediction inputs and flexible guidance
- JiT — clean-image prediction in pixel space
- Pixel MeanFlow — one-step pixels through a denoised field
- ML 09 · Autoregressive Vision: Pixels, Tokens and Scales
1 optional page: Paper guides 1
- ML 10 · Foundation Models and Vision–Language Learning
1 optional page: Paper guides 1
- ML 11 · Evaluation, Efficiency and Deployment
1 optional page: Paper guides 1
Deep Learning for Computer Vision
14 main chapters
DLCV 01 · Deep Learning for Image Processing and Computer Vision3 sub-chapters
3 optional pages: Supplementary notes 2 · Paper guides 1
- DLCV 01A · Training Deep Vision Models in Practice
6 optional pages: Supplementary notes 4 · Paper guides 2
- DLCV 01B · Building Blocks and How to Compose Them: Classification and Restoration Networks
9 optional pages: Supplementary notes 5 · Paper guides 4
- DLCV 01C · Model Compression and Efficient Inference
8 optional pages: Supplementary notes 6 · Paper guides 2
DLCV 02 · Deep Image Classification1 sub-chapter
10 optional pages: Supplementary notes 9 · Paper guides 1
Supplementary notes- A worked macro-versus-micro example
- Cohen’s kappa: agreement beyond chance
- Multi-label metrics in detail
- Evaluation benchmarks in detail
- How reliable are ImageNet numbers?
- Why residual networks train: the unrolled gradient
- Two more CNN designs: DenseNet and RegNet
- Squeeze-and-excitation: attention inside a CNN
- Contrastive learning in detail: SimCLR and MoCo
- DLCV 02A · Face Recognition and Metric Learning
9 optional pages: Supplementary notes 6 · Paper guides 3
Supplementary notes- Aligning a face: the least-squares similarity transform
- How large must the scale s be?
- Margin losses side by side: one formula and its edge cases
- Beyond one center per class: sub-centers, quality-aware margins and Partial FC
- Measuring very small error rates
- Before deep learning: eigenfaces and Fisherfaces
DLCV 03 · Object Detection: From R-CNN to DETR2 sub-chapters
6 optional pages: Supplementary notes 5 · Paper guides 1
- DLCV 03A · One-Stage Detection, Read Through YOLOv4
24 optional pages: Supplementary notes 2 · Paper guides 22
Paper guides- YOLO v1: detection as a single regression
- YOLOv2 and YOLO9000: anchors, priors and joint training
- YOLOv3: three scales and Darknet-53
- SSD: multi-scale default boxes
- RetinaNet and the focal loss
- FCOS: per-pixel anchor-free detection
- CenterNet: objects as points
- YOLOR: explicit features and learned implicit representations
- YOLOX: anchor-free YOLO with SimOTA
- YOLOv6: a re-parameterized industrial detector
- YOLOv7: a trainable bag of freebies
- YOLOv9: programmable gradient information
- YOLOv10: NMS-free training with dual assignments
- YOLOv12: an attention-centric YOLO
- YOLOv13: hypergraph correlation
- YOLO26: a DFL-free end-to-end YOLO
- RT-DETR: a real-time DETR
- RT-DETRv2: a bag of freebies for RT-DETR
- D-FINE: boxes as refined distributions
- DEIM: dense one-to-one matching
- RF-DETR: architecture search on a DINOv2 backbone
- DEIMv2: DINOv3 features for real-time DETRs
- DLCV 03B · 3D Object Detection: LiDAR, Cameras and Bird's-Eye View
8 optional pages: Supplementary notes 5 · Paper guides 3
- DLCV 04 · Image Segmentation: Semantic, Instance and Panoptic
6 optional pages: Supplementary notes 4 · Paper guides 2
- DLCV 05 · Human Pose and Keypoint Estimation
8 optional pages: Supplementary notes 5 · Paper guides 3
- DLCV 06 · Text in Images: OCR and Document Understanding
9 optional pages: Supplementary notes 5 · Paper guides 4
- DLCV 07 · Object Tracking: From Single-Object to Multi-Object
5 optional pages: Supplementary notes 4 · Paper guides 1
- DLCV 08 · Video Understanding and Action Recognition
7 optional pages: Supplementary notes 4 · Paper guides 3
DLCV 09 · Deep Image Restoration: Low-Level Vision3 sub-chapters
24 optional pages: Supplementary notes 11 · Paper guides 13
Supplementary notes- Blind super-resolution: conditioning on and estimating the kernel
- Plug-and-play priors and half-quadratic splitting
- Super-resolution milestones, families and benchmarks
- Classical and learned blind-deblurring methods
- All-in-one methods in more detail
- Human-rated IQA datasets
- PIQE and the Perceptual Index
- How the learned NR assessors work
- CLIP-IQA and Q-Align scores
- Fitting the Bradley–Terry model
- Metric-by-metric summary table
Paper guides- Restormer: global channel relationships without pixel-wise attention
- AirNet: infer the degradation before conditioning restoration
- TransWeather: learned weather queries for a shared restoration model
- PromptIR: input-conditioned feature prompts, without language instructions
- AdaIR: adaptively separate and exchange frequency information
- DA-CLIP: separate content guidance from degradation guidance
- InstructIR: route image features with a human instruction
- AutoDIR: diagnose degradation, guide diffusion, then correct structure
- AgenticIR: plan, inspect and revise a sequence of restoration tools
- 4KAgent: select restoration experts for a configured fidelity goal
- PromptFix: combine instructions, visual diagnosis and high-frequency preservation
- Deep Image Prior: the network as the prior
- Survey papers in detail
DLCV 09A · Non-Blind Image Super-Resolution3 sub-chapters
6 optional pages: Supplementary notes 6
- DLCV 09A.1 · Blind and Real-World Image Super-Resolution
14 optional pages: Supplementary notes 3 · Paper guides 11
Paper guides- SRMD: one network for many known degradations (CVPR 2018)
- IKC: iterative kernel correction (CVPR 2019)
- KernelGAN: kernel estimation with an internal GAN (NeurIPS 2019)
- DAN: unfolding the alternating optimization (NeurIPS 2020)
- DASR: contrastive degradation representations (CVPR 2021)
- BSRGAN: a practical degradation model (ICCV 2021)
- Real-ESRGAN: high-order degradation with pure synthetic data (ICCVW 2021)
- StableSR: a time-aware encoder for a frozen diffusion prior (IJCV 2024)
- DiffBIR: remove the degradation, then regenerate (ECCV 2024)
- SeeSR: degradation-aware semantic prompts (CVPR 2024)
- SUPIR: scaling up the prior, the data and the text (CVPR 2024)
- DLCV 09A.2 · One-Step Diffusion Super-Resolution
10 optional pages: Supplementary notes 2 · Paper guides 8
Paper guides- ResShift: a short restoration-conditioned chain (NeurIPS 2023)
- SinSR: distilling a deterministic image-conditioned map (CVPR 2024)
- OSEDiff: LR-latent start and variational score regularization (NeurIPS 2024)
- S3Diff: degradation-guided low-rank adaptation
- TSD-SR: target-aware score regularization (CVPR 2025)
- InvSR: a learned image-conditioned start (CVPR 2025)
- PiSA-SR: separate pixel and semantic adaptation (CVPR 2025)
- AddSR: prediction-refined adversarial restoration distillation
- DLCV 09B · Generative Image Restoration: Sampling a Plausible Answer
9 optional pages: Supplementary notes 4 · Paper guides 5
Paper guides- Blau and Michaeli, “The Perception-Distortion Tradeoff” (CVPR 2018)
- Chung et al., “Diffusion Posterior Sampling for General Noisy Inverse Problems” (ICLR 2023)
- DDNM: preserve measured components and generate the null space
- BlindDPS: infer an image and its degradation together
- Li et al., “Diffusion Models for Image Restoration and Enhancement: A Comprehensive Survey” (IJCV 2025)
- DLCV 09C · Low-Light Image Enhancement
27 optional pages: Supplementary notes 14 · Paper guides 13
Supplementary notes- CLAHE: equalizing locally without amplifying noise
- More deep LLIE methods
- Counting resolution, arithmetic and memory
- A bilateral-grid experiment
- Implementing and checking trilinear LUTs
- Fitting and slicing a bilateral grid in detail
- Learnable LUT variants in detail
- Structural re-parameterization step by step
- Understanding no-reference low-light metrics
- Efficient network building blocks
- Banding, halos and temporal stability
- Bilateral grids and color-table mathematics
- From retouching to low light
- A lightness-order exercise
Paper guides- KinD: repair reflectance while controlling illumination
- Retinex-Net: decompose, adjust and denoise
- EnlightenGAN: unpaired appearance learning with content constraints
- Zero-DCE and Zero-DCE++: learn curves without reference images
- RUAS: search the priors inside an unrolled enhancement model
- SCI: train with calibration, deploy one illumination block
- Retinexformer: illumination guides restoration
- IAT: local affine maps and a global camera-style transform
- Multinex: fuse analytic priors into a Retinex residual
- CPGA-Net: channel priors, gamma estimation and learned fusion
- CPGA-Net+: attention, BDSF and stronger variants
- HDRNet: deep bilateral learning for real-time image enhancement
- Image-adaptive 3D LUTs (Zeng et al., TPAMI 2022)
- DLCV 10 · Image Fusion: Complementary Images, Useful Outputs
17 optional pages: Supplementary notes 6 · Paper guides 11
Paper guides- Exposure fusion (Mertens, Kautz and Van Reeth)
- VIFB: a visible and infrared image fusion benchmark
- The infrared–visible fusion survey of Liu et al. (TPAMI 2025)
- DenseFuse: learn reconstruction, choose fusion in feature space
- DDFM: a diffusion prior with likelihood rectification
- FusionGAN: thermal intensity with an adversarial visible-image prior
- IFCNN: supervised fusion with a variable number of inputs
- U2Fusion: one unsupervised model across fusion tasks
- SwinFusion: attention within and across the sources
- SeAFusion: let semantic segmentation influence the fused image
- TarDAL: fuse for both appearance and object detection
DLCV 11 · 3D Reconstruction: From Multi-View Geometry to NeRF and Gaussian Splatting1 sub-chapter
10 optional pages: Supplementary notes 7 · Paper guides 3
DLCV 12 · Image Generation: From GANs to Text-to-Image Systems2 sub-chapters
7 optional pages: Supplementary notes 4 · Paper guides 3
- DLCV 12A · Video Generation
6 optional pages: Supplementary notes 4 · Paper guides 2
- DLCV 12B · One-Step Generation
9 optional pages: Supplementary notes 2 · Paper guides 7
Paper guides- Progressive Distillation: two teacher moves in one student move (ICLR 2022)
- Consistency Models: the same endpoint along a path (ICML 2023)
- InstaFlow: reflow, then endpoint distillation (ICLR 2024)
- Latent Consistency Models: endpoints in a guided latent flow (2023)
- Adversarial Diffusion Distillation: a teacher and a discriminator (ECCV 2024)
- DMD: matching a teacher-defined distribution (CVPR 2024)
- DMD2: tracking the student and using real data (NeurIPS 2024)
- DLCV 14 · Robustness: Distribution Shift, Adversarial Examples and Anomalies
9 optional pages: Supplementary notes 6 · Paper guides 3
Optimization
8 main chapters
- OPT 05 · Approximation, Statistical Estimation and Experiment Design
1 optional page: Supplementary notes 1
Information Theory
8 main chapters
- IT 07 · Statistical Evidence, Maximum Entropy and Growth
1 optional page: Supplementary notes 1