Vision Model Feature Extraction
15 repos
Tools and implementations for extracting and encoding visual features from images using pre-trained perception models, with applications in zero-shot image classification and feature-based image retrieval. The cluster includes optimized encoder architectures and tiling strategies for processing large-scale image datasets efficiently, combining both inference frameworks and research implementations exploring model architecture variations and performance tuning.
Jupyter Notebook · 1
perception-encoder ·143
zero-shot-image-classification ·97
image-feature-extraction ·46
