Non-Max Suppression (NMS): Computer Vision Mastery Guide (2026)

Progress 4 / 8

📑 Table of Contents

Introduction to Computer Vision

1 What is Computer Vision? Use Cases, Skills, and Roadmap
2 Images as Tensors: Pixels, Channels, and Batch Dimensions
3 Problem Framing: Classification vs Detection vs Segmentation
4 Dataset Strategy: Diversity, Labeling Rules, and Leakage Prevention
5 Classical CV vs Deep Learning: A Practical Decision Guide
6 CV System Design Basics: Latency, Accuracy, and Cost
7 Data → Model → Output → Feedback Loop (Production Workflow)
8 Mini Project Blueprint: Ship Your First Image Classifier

Image Fundamentals & Digital Image Processing

9 Sampling & Quantization: Resolution, Bit-Depth, and Visual Artifacts
10 Color Spaces: RGB vs HSV vs LAB (When to Switch)
11 Convolution Intuition: Kernels, Filters, and Feature Detection
12 Histograms & Contrast: Equalization and CLAHE
13 Noise Models & Denoising: Gaussian, Salt-and-Pepper, Motion Blur
14 Geometric Transforms: Affine vs Perspective (Homography)
15 Edge Detection Deep Dive: Canny Tuning That Actually Works
16 Frequency Domain Basics: FFT and Filtering

Image Preprocessing & Augmentation

17 Normalization: Scale, Mean/Std, and Training Stability
18 Resizing Strategies: Aspect Ratio, Padding, Letterboxing
19 Augmentation Fundamentals: What Helps vs What Hurts
20 MixUp, CutMix, Mosaic: Modern Augmentations Explained
21 Augmentation for Detection & Segmentation: Label-Safe Transforms
22 Lighting Robustness: Color Augmentations That Simulate Reality
23 Preprocessing for Low-Light and Noisy Cameras
24 Production Preprocessing Checklist: Avoid Silent Accuracy Drops

Feature Extraction & Classical Computer Vision

25 Keypoints & Descriptors: The Core Idea Behind Matching
26 HOG Features: Why Gradients Capture Shape
27 ORB vs SIFT vs SURF: Trade-offs for Real Projects
28 Template Matching: High-Speed Detection in Controlled Scenes
29 Corner Detection: Harris and Shi-Tomasi in Practice
30 Optical Flow: Motion Tracking Between Frames
31 Classical Pipeline: Features → Matching → Geometry → Decision
32 Decision Guide: Classical vs Deep Learning in 2026

Convolutional Neural Networks (CNNs)

33 CNN Layers Explained: Convolution, Pooling, Activation, BatchNorm
34 Backprop in CNNs: A Practical Intuition
35 Training a Classifier: Losses, Optimizers, and Schedules
36 Overfitting: Detection and Fixes That Actually Work
37 Grad-CAM and Feature Map Visualization
38 Batch Size vs Learning Rate: Stability and Speed
39 From Scratch vs Pretrained CNNs: A Cost/Benefit Guide
40 Mini Project Checklist: Build a CNN Classifier End-to-End

Advanced CNN Architectures

41 VGG, Inception, ResNet: Ideas That Shaped Modern CNNs
42 Residual Connections Deep Dive
43 EfficientNet and Compound Scaling
44 MobileNet for Edge Vision
45 Normalization Choices in Vision
46 Regularization for Vision Nets
47 Designing Custom CNNs
48 Benchmarking Architectures for Production

Object Detection & Localization

49 Detection Basics: Boxes, Anchors, IoU
50 Two-Stage Detectors: Faster R-CNN
51 One-Stage Detectors: YOLO/SSD
52 Non-Max Suppression (NMS)
53 Small Object Detection
54 Training Detectors
55 Detection Evaluation: mAP
56 Deploying Detectors in Streams

Image Segmentation Techniques

57 Segmentation Types: Semantic vs Instance vs Panoptic
58 U-Net Architecture
59 Segmentation Losses
60 Mask R-CNN Overview
61 Mask Post-Processing
62 Annotation Strategy for Masks
63 Segmentation Metrics
64 Real-Time Segmentation

Transfer Learning in Computer Vision

65 Why Transfer Learning Works
66 Feature Extraction vs Fine-Tuning
67 Domain Shift Problems
68 Layer Freezing Recipe
69 Fine-Tuning With Limited Data
70 Transfer for Detection/Segmentation
71 Self-Supervised Pretraining
72 Common Pitfalls

Vision Transformers (ViT) & Modern Architectures

73 ViT Basics: Patches and Tokens
74 ViT vs CNN
75 Hybrid CNN-Transformer Models
76 Swin Transformer
77 Attention Maps: Interpretability
78 Training ViTs
79 Transformers for Detection: DETR
80 Backbone Selection in 2026

Face Recognition & Biometric Systems

81 Face Recognition Pipeline
82 Face Detectors
83 Face Embeddings and Metric Learning
84 Threshold Tuning: FAR/FRR
85 Liveness Detection
86 Bias and Fairness
87 Privacy & Compliance
88 System Design: Enrollment to Monitoring

Video Analysis & Action Recognition

89 Video Fundamentals
90 Tracking-by-Detection
91 Optical Flow
92 3D CNNs for Action Recognition
93 Temporal Models: LSTM vs Transformers
94 Real-Time Video Pipelines
95 Video Evaluation
96 Surveillance Analytics Case Study

Model Evaluation & Metrics in Computer Vision

97 Classification Metrics
98 Confusion Matrix Reading
99 Detection Metrics: mAP
100 Segmentation Metrics
101 Calibration
102 Robustness Testing
103 Error Analysis Workflow
104 Online Monitoring

Computer Vision Deployment & MLOps

105 From Notebook to API
106 Export Formats
107 TensorRT Optimization
108 Docker for Inference
109 Batching vs Real-Time
110 Monitoring CV Models
111 CI/CD for Models
112 Edge Deployment

Advanced Topics & Research in Computer Vision

113 GANs vs Diffusion
114 Diffusion Intuition
115 Vision-Language Models
116 Self-Supervised Vision
117 Synthetic Data
118 Adversarial Robustness
119 Privacy-Preserving Vision
120 Research Trends 2026

Non-Max Suppression (NMS)

Intermediate Topic 4 of 8