Glossario IA
Il dizionario completo dell'Intelligenza Artificiale
Convolution Layers
Mathematical operations applying filters to extract local features from images.
Pooling Layers
Dimensionality reduction operations to decrease the size of feature maps.
Classic CNN Architectures
Fundamental models like LeNet, AlexNet, and VGG that established the foundation of modern CNNs.
Deep and Residual CNNs
Advanced architectures like ResNet and DenseNet enabling the training of very deep networks.
CNN for Semantic Segmentation
Specialized applications like U-Net for pixel-by-pixel classification of images.
CNN for Object Detection
Systems like YOLO and R-CNN to locate and classify multiple objects in an image.
3D and Spatio-temporal CNNs
Networks processing volumetric or video data with three-dimensional convolutions.
Transfer Learning with CNN
Techniques for reusing pre-trained models to accelerate learning on new tasks.
CNN for Image Generation
Applications in GANs and VAEs to create new realistic synthetic images.
CNN Optimization and Regularization
Techniques like dropout, batch normalization, and data augmentation to improve performance.
CNN for Computer Vision
Integrated applications for scene recognition, depth estimation, and object tracking.
CNN for Medical Analysis
Specialized applications for diagnosis from medical images such as MRI and scans.
Lightweight and Mobile CNNs
Optimized architectures like MobileNet and SqueezeNet for resource-limited devices.
Attention Mechanisms in CNN
Integration of attention mechanisms to focus on relevant regions of images.
Multi-task CNN
Architectures simultaneously performing multiple computer vision tasks.