Differences

This shows you the differences between two versions of the page.

--- ai:methods [2022/04/07 22:07]
cyril import private
+++ — (current)
@@ Line 1: / Line 1: @@
-====== Catalog of methods in AI/vision/robotics ======
-This is a classification of techniques and algorithms, in order to give a broad view of solutions available to deal with classical problems.
-Other pages contain some more details about, which can be seen as memos with references to find more information (with the origin paper when possible).
-===== Learning =====
-==== Supervised learning ====
-=== Classification ===
-  * __MLP (Multi Layers Perceptron)__ - //PMC (Perceptron multicouches)//
-    * gradient backpropagation - //rétropropagation du gradient//
-      * stochastic
-      * with inertia
-      * simulated annealing - //recuit simulé//
-    * newton (second order)
-  * __RBFNN (Radial Basis Functions Neural Networks)__
-    * k-means then gradient descent
-    * incremental addition of neurons then exact method
-  * __SVM (Support Vectors Machine)__
-  * __Decision tree__ - //arbre de décision//
-    * ID3 (based on entropy)
-  * __k-nearest neighbors__ - //k plus proches voisins//
-  * __Boosting__
-    * Boosting by majority
-    * AdaBoost (ADAptive BOOSTing)
-      * Discrete AdaBoost
-      * Real AdaBoost
-      * Gentle AdaBoost
-      * LogitBoost
-      * Probabilistic AdaBoost
-      * FloatBoost
-      * AdaBoost.Reg
-      * Multiclass AdaBoost.M1
-      * Multiclass AdaBoost.M2
-      * Multilabel AdaBoost.MR
-      * Multilabel AdaBoost.MH
-      * Multiclass AdaBoost.MO
-      * Multiclass AdaBoost.OC
-      * Multiclass AdaBoost.ECC
-      * GrPloss
-      * BoostMA
-      * AdaBoost.M1W
-      * SAMME (Stagewise Additive Modeling using a Multi-class Exponential loss function)
-      * GAMBLE (Gentle Adaptive Multiclass Boosting Learning)
-    * UBoost
-    * LPBoost (Linear Programming BOOSTing)
-    * TotalBoost (TOTALly corrective BOOSTing)
-    * RotBoost
-    * alphaBoost
-    * MILBoost (Multiple Instance Learning BOOSting)
-    * CGBoost (Conjugate Gradient BOOSTing)
-    * Bootstrap Aggregating
-  * __Cascades of detectors__ [[classification#cascades_of_detectors|[+] ]]
-  * __Trees of detectors__
-=== Regression ===
-  * __MLP (Multi Layers Perceptron)__
-  * __RBFNN (Radial Basis Functions Neural Network)__
-  * __SVR (Support Vectors Regressor)__
-=== Pattern recognition ===
-  * __Viola-Jones Detector__ [[pattern-recognition#viola-jones_detector|[+] ]]
-    * with Extended Set of Haar features
-    * Stumps or CART trees
-    * Rotation Invariant
-    * **Multiview**
-      * Parallel Cascades
-      * Pyramid Cascade
-      * Tree Cascade
-      * Vector Boosting
-==== Unsupervised learning ====
-=== Vector quantization / Clustering ===
-  * Sequential leader
-  * k-means - //k-moyennes//
-  * GNG (Growing Neural Gas)
-  * Auto-organizing maps (Kohonen) - //cartes auto-organisatrices de Kohonen//
-==== Reinforcement learning ====
-  * __MDP (Markov Decision Processes)__
-    * Q-learning
-    * Value iteration
-    * Policy iteration
-===== Planification =====
-==== Symbolic ====
-=== State space search ===
-  * A*
-    * WA*
-    * IDA*
-  * Dijkstra
-=== Logics ===
-  * __GraphPlan__
-    * Stan
-    * IPP
-    * SGP
-  * __SATplan (SATisfiability PLANning)__
-==== Others ====
-  * __Genetic algorithms__ - //algorithmes génétiques//
-  * __Ant colonies__ - //colonies de fourmis//
-==== Specific ====
-=== Path planning ===
-  * __Configurations space__
-  * __Potential fields__
-===== Perception =====
-==== Vision ====
-=== Color Quantization ===
-  * RGB cone
-  * YUV polygon
-  * HSV rectangle
-  * Lab
-=== Image segmentation ===
-  * Floodfill - //inondation//
-  * Watershed - //lignes de partage des eaux//
-=== Filters ===
-  * **Anti-noise (smoothing)**
-    * Median Filter - //filtre médian//
-    * Vector Median Filter
-    * Kuwahara filter
-    * Peer Group Filtering
-    * Anisotropic Filtering
-  * **Gradient** [[vision#gradient|[+] ]]
-    * Prewitt [[vision#prewitt|[+] ]]
-    * Roberts [[vision#roberts|[+] ]]
-    * Sobel [[vision#sobel|[+] ]]
-    * Laplace [[vision#laplace|[+] ]]
-    * Scharr [[vision#scharr|[+] ]]
-  * **Morphological**
-    * dilation - //dilatation//
-    * erosion - //érosion//
-    * opening - //ouverture//
-    * closing - //fermeture//
-=== Edge detection ===
-  * Canny detector
-    * Canny-Deriche
-=== Pattern recognition ===
-  * Mean-Square Regression - //Régression aux moindres carrés//
-  * __Hough Transforms__ [[pattern-recognition#hough_transforms|[+] ]]
-    * Standard Hough Transform
-    * Randomized Hough Transform
-    * Connective Randomized Hough Transform
-    * Combinatorial Hough Transform
-    * Adaptive Hough Transform
-    * Probabilistic Hough Transform
-    * Adaptive Probabilistic Hough Transform
-    * Progressive Probabilistic Hough Transform
-    * Hierarchical Hough Transform
-    * Sampling Hough Transform
-    * Generalized Hough Transform
-  * UpWrite method
-  * Curvogram
-  * **Shape Descriptors**
-    * Ankerst's Shape histograms
-  * Chamfer matching
-    * Contour Likelihood Measurement
-=== Tracking ===
-  * Kalman Filter
-    * Generalized Kalman Filter
-  * Correlation Tracking
-    * ACA (Area Correlation Algorithm)
-  * KLT Tracker (Kanade-Lucas-Tomasi)
-  * IPAN Tracker
-  * MeanShift
-  * **Features detection**
-    * Harris detector
-    * Susan detector
-    * Multiresolution Contrast detector
-==== Sensors fusion ====
-  * Kalman filter
-  * Particles filter (bayesian network) - //filtrage particulaire//