The iron ML notebook
  • The iron data science notebook
  • ML & Data Science
    • Frequent Questions
      • Discriminative vs Generative models
      • Supervised vs Unsupervised learning
      • Batch vs Online Learning
      • Instance-based vs Model-based Learning
      • Bias-Variance Tradeoff
      • Probability vs Likelihood
      • Covariance vs Correlation Matrix
      • Precision vs Recall
      • How does a ROC curve work?
      • Ridge vs Lasso
      • Anomaly detection methods
      • How to deal with imbalanced datasets?
      • What is "Statistically Significant"?
      • Recommendation systems methods
    • Statistics
      • The basics
      • Distributions
      • Sampling
      • IQR
      • Z-score
      • F-statistic
      • Outliers
      • The bayesian basis
      • Statistic vs Parameter
      • Markov Monte Carlo Chain
    • ML Techniques
      • Pre-process
        • PCA
      • Loss functions
      • Regularization
      • Optimization
      • Metrics
        • Distance measures
      • Activation Functions
      • Selection functions
      • Feature Normalization
      • Cross-validation
      • Hyperparameter tuning
      • Ensemble methods
      • Hard negative mining
      • ML Serving
        • Quantization
        • Kernel Auto-Tuning
        • NVIDIA TensorRT vs ONNX Runtime
    • Machine Learning Algorithms
      • Supervised Learning
        • Support Vector Machines
        • Adaptative boosting
        • Gradient boosting
        • Regression algorithms
          • Linear Regression
          • Lasso regression
          • Multi Layer Perceptron
        • Classification algorithms
          • Perceptron
          • Logistic Regression
          • Multilayer Perceptron
          • kNN
          • Naive Bayes
          • Decision Trees
          • Random Forest
          • Gradient Boosted Trees
      • Unsupervised learning
        • Clustering
          • Clustering metrics
          • kMeans
          • Gaussian Mixture Model
          • Hierarchical clustering
          • DBSCAN
      • Cameras
        • Intrinsic and extrinsic parameters
    • Computer Vision
      • Object Detection
        • Two-Stage detectors
          • Traditional Detection Models
          • R-CNN
          • Fast R-CNN
          • Faster R-CNN
        • One-Stage detectors
          • YOLO
          • YOLO v2
          • YOLO v3
          • YOLOX
        • Techniques
          • NMS
          • ROI Pooling
        • Metrics
          • Objectness Score
          • Coco Metrics
          • IoU
      • MOT
        • SORT
        • Deep SORT
  • Related Topics
    • Intro
    • Python
      • Global Interpreter Lock (GIL)
      • Mutability
      • AsyncIO
    • SQL
    • Combinatorics
    • Data Engineering Questions
    • Distributed computation
      • About threads & processes
      • REST vs gRPC
  • Algorithms & data structures
    • Array
      • Online Stock Span
      • Two Sum
      • Best time to by and sell stock
      • Rank word combination
      • Largest subarray with zero sum
    • Binary
      • Sum of Two Integers
    • Tree
      • Maximum Depth of Binary Tree
      • Same Tree
      • Invert/Flip Binary Tree
      • Binary Tree Paths
      • Binary Tree Maximum Path Sum
    • Matrix
      • Set Matrix Zeroes
    • Linked List
      • Reverse Linked List
      • Detect Cycle
      • Merge Two Sorted Lists
      • Merge k Sorted Lists
    • String
      • Longest Substring Without Repeating Characters
      • Longest Repeating Character Replacement
      • Minimum Window Substring
    • Interval
    • Graph
    • Heap
    • Dynamic Programming
      • Fibonacci
      • Grid Traveler
      • Can Sum
      • How Sum
      • Best Sum
      • Can Construct
      • Count Construct
      • All Construct
      • Climbing Stairs
Powered by GitBook
On this page

Was this helpful?

  1. ML & Data Science
  2. Computer Vision
  3. Object Detection
  4. One-Stage detectors

YOLOX

  • Released in 2021, from the YOLO v3 implementation

  • Anchor-free architecture

  • Multi positives: for compensating the large imbalance that the abserce of anchors produces, added center sampling with 3 x 3 areas as positives.

  • Decoupled head: separates classification confidence and localization accuracy into two heads (classification and regression).

  • Advanced label assignment: to avoid the problem the GT label assignment could have ambiguities when the boxes of multiple objects overlap.

  • Strong augmentations: uses MixUP and Mosaic, avoiding ImageNet pretraining.

Center sampling

In anchor-free object detection models, predictions are typically made at various points (or pixels) on a feature map. During training, these models need to determine which points should be classified as positive samples (those belonging to an object) and which should be negatives (background). This selection process is crucial for learning accurate object localization and classification.

Center sampling focuses on assigning positive samples only to points that are closer to the center of an object's bounding box, rather than to every point within the bounding box. Here’s how it generally works:

  1. Bounding Box Center Region: A region around the center of each object's bounding box is defined. This region is usually a smaller, central area of the entire bounding box. The size of this center region can be fixed or dynamically adjusted based on the size of the object.

  2. Sample Assignment: During training, only the points (or pixels) within this center region are considered as potential positive samples for that object. Points outside this center region are considered negative, even if they lie within the bounding box. This reduces the ambiguity of sample selection and helps the model focus on more informative areas.

  3. Dynamic Sampling: Some implementations of center sampling allow the size of the central region to adapt based on the size of the object. For example, the central region might be a fraction of the width and height of the bounding box, ensuring that the model can handle objects of various sizes.

PreviousYOLO v3NextTechniques

Last updated 9 months ago

Was this helpful?