목록전체 글 (80)
studio.heelab
Lecture: https://www.youtube.com/watch?v=aVJy4O5TOk8&list=PLoROMvodv4rOmsNzYBMe0gJY2XS8AQg16&index=6 1. Building Blocks of CNNsComponents of CNNs Convolutional Layers: Filters (kernels) slide across the image, producing an activation map through dot products. The output depth must match the number of filters used, and the filter depth of the next layer must equal the preceding output's depth. Po..
Lecture: https://www.youtube.com/watch?v=f3g1zGdxptI&list=PLoROMvodv4rOmsNzYBMe0gJY2XS8AQg16&index=5 1. DL Fundamentals Recapmage Classification and Linear Classifiers: A method of defining input as a tensor and predicting class scores through a weight matrix $W$.Loss Functions: Measures how well the model fits the data using functions such as Softmax or SVM.Optimization: Utilizes algorithms lik..
Lecture: https://www.youtube.com/watch?v=25zD5qJHYsk&list=PLoROMvodv4rOmsNzYBMe0gJY2XS8AQg16&index=4 1. Basic Structure of Neural Networks Neural networks: the original linear classifier -> 2layers Multi-layer Structure: Beyond a single linear layer (W x X), neural networks are constructed by stacking multiple layers, such as W_2 x max(0, W_1 x X). Hidden Layers: Intermediate neurons learn spe..
Lecture: https://www.youtube.com/watch?v=dyNGd06MWn4&list=PLoROMvodv4rOmsNzYBMe0gJY2XS8AQg16&index=3 1. RegularizationAdding an extra term to the loss function to prevent the model from overfitting to the training data. The goal is to improve performance on unseen test data (generalization) even if it slightly reduces training performance. hyperparameter (Lambda): A value that controls the stren..
Lecture: https://www.youtube.com/watch?v=pdqofxJeBN8&list=PLoROMvodv4rOmsNzYBMe0gJY2XS8AQg16&index=2 1. Image ClassificationThe problem: Semantic GapAn image is a tensor of integers between [0,255] Challenges-Viewpoint variation -Illumination -Background Clutter -Occlusion -Deformation-Intraclass variation An image classifierdef classify_image(image): #Some magic here: return class_label Fin..
Lecture: https://www.youtube.com/playlist?list=PLoROMvodv4rOmsNzYBMe0gJY2XS8AQg16 Stanford CS231N Deep Learning for Computer Vision I 2025Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving car...www.youtube.com Lecture 1: IntroductionAgenda-A breif history of CV and DL 1. The Essence and Histor..