This project is a computer vision dataset and image annotation repository designed for training and evaluating machine learning models. It provides a large collection of labeled images, serving as an object detection benchmark and a source of pixel-level segmentation data. The repository distinguishes itself as a multimodal visual dataset by pairing images with synchronized voice, text, and mouse traces to support narrative understanding. It further enables the analysis of model fairness through the inclusion of demographic attributes and exhaustive annotations. The dataset covers a broad ra
BBox-Label-Tool is a web-based utility designed for labeling image collections and defining spatial object boundaries to support supervised machine learning tasks. It provides an interface for drawing rectangular bounding boxes on images, allowing users to record coordinate data for object detection and visual recognition datasets. The tool operates entirely within the browser, utilizing local file processing to read images directly from the user's system without requiring data uploads. It maintains annotation records through browser-based storage, ensuring that spatial data persists across p
OpenSeadragon is a JavaScript library and tiled image rendering engine designed for high-resolution image viewing. It functions as a deep zoom image viewer that renders massive images using a tiled pyramid approach, enabling smooth panning and zooming without requiring the full image file to be loaded. The project distinguishes itself through broad support for standardized image retrieval protocols, including the International Image Interoperability Framework (IIIF), IIPImage, Iris, and OpenStreetMap. It provides a hardware-accelerated rendering layer via WebGL to apply real-time filters and
A multi-purpose Video Labeling GUI in Python with integrated SOTA detector and tracker