Dataset Hub

The UW‑Bench dataset, created by the School of Microelectronics and Communication Engineering at Chongqing University, focuses on urban flood detection, containing 7,677 annotated images sourced from surveillance cameras and handheld devices. The dataset was collected under various adverse conditions such as low light, strong reflections, and clear water, aiming to improve model generalization in real‑world applications. Manual annotation ensures data quality, suitable for enhancing accuracy and efficiency of urban flood detection.

Office‑31 consists of 31 office‑object categories, Office‑Home contains 65 everyday‑object categories, VisDA‑2017 is a dataset for visual domain adaptation challenges, and DomainNet is a large‑scale multi‑domain image dataset.

A dataset containing over 500,000 mineral images, each labeled, sourced from mindat.org. The dataset includes two CSV files that store image URLs and cleaned label information.

The IMDb‑Face dataset is used for face recognition and contains facial images gathered from IMDb. The Megaface dataset is a large‑scale face recognition benchmark comprising multiple subsets for various recognition tasks.

dSprites is a 2D shape dataset generated from six basic independent latent factors (color, shape, scale, rotation, x‑position, and y‑position) for evaluating the disentanglement properties of unsupervised learning methods. The dataset contains 737,280 images, each representing a unique combination of these latent factors.