19,997 machine learning datasets
19,997 dataset results
RealWorldQA is a benchmark designed to evaluate the real-world spatial understanding capabilities of multimodal AI models. It assesses how well these models comprehend physical environments. The benchmark consists of over 700 images, each accompanied by a question and a verifiable answer. These images are drawn from various real-world scenarios, including those captured from vehicles. The goal is to advance AI models' understanding of our physical world.
<img src="https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F12294787%2F2e9b3b5a8f236aab36655b4a0db4e311%2Foverview.jpg?generation=1718265309709943&alt=media" alt="drawing" style="width:700px;"/>
We construct a fine-grained video-text dataset with 12K annotated high-resolution videos (~400k clips). The annotation of this dataset is inspired by the video script. If we want to make a video, we have to first write a script to organize how to shoot the scenes in the videos. To shoot a scene, we need to decide the content, shot type (medium shot, close-up, etc), and how the camera moves (panning, tilting, etc). Therefore, we extend video captioning to video scripting by annotating the videos in the format of video scripts. Different from the previous video-text datasets, we densely annotate the entire videos without discarding any scenes and each scene has a caption with ~145 words. Besides the vision modality, we transcribe the voice-over into text and put it along with the video title to give more background information for annotating the videos.
The OpenEQA dataset is a significant contribution in the field of Embodied Question Answering (EQA). Let me provide you with some details:
CloudSEN12 is a LARGE dataset (~1 TB) for cloud semantic understanding that consists of 49,400 image patches (IP) that are evenly spread throughout all continents except Antarctica. Each IP covers 5090 x 5090 meters and contains data from Sentinel-2 levels 1C and 2A, hand-crafted annotations of thick and thin clouds and cloud shadows, Sentinel-1 Synthetic Aperture Radar (SAR), digital elevation model, surface water occurrence, land cover classes, and cloud mask results from six cutting-edge cloud detection algorithms.
SuperBench is a comprehensive evaluation system for large language models that includes five benchmark datasets: ExtremeGLUE for semantics, CodeBench for code, AlignBench for alignment, AgentBench for intelligent agents, and SafetyBench for safety. These benchmarks cover a wide range of tasks and dimensions to assess the overall capabilities of large language models. The SuperBench team aims to provide objective and scientific evaluation standards for large models to promote their healthy development in terms of technology, applications, and ecosystem.
This dataset comprises the raw data of four three-dimensional seismic surveys acquired in Canadian mining camps. They include first-break picks in the trace headers. Please see all details on the github repo.
The COCONut dataset is a modernized segmentation dataset that builds upon the established COCO benchmark. It aims to address the limitations of the original COCO segmentation annotations by enhancing annotation quality and expanding the dataset to encompass a larger number of images with high-quality masks. COCONut harmonizes segmentation annotations across semantic, instance, and panoptic segmentation tasks, providing meticulously crafted masks for improved accuracy and consistency. It includes 383K images with more than 5.18M panoptic masks, making it a large-scale universal segmentation dataset verified by human raters [T1], [T6].
Click to add a brief description of the dataset (Markdown and LaTeX enabled).
The Arena-Hard benchmark is a high-quality benchmarking tool for Language Learning Models (LLMs) developed by LMSYS Org¹. It was designed to address the limitations of traditional benchmarks, which are often static or close-ended¹.
ShipRSImageNet is a large-scale fine-grainted dataset for ship detection in high-resolution optical remote sensing images. The dataset contains 3,435 images from various sensors, satellite platforms, locations, and seasons. Each image is around 930×930 pixels and contains ships with different scales, orientations, and aspect ratios. The images are annotated by experts in satellite image interpretation, categorized into 50 object categories images. The fully annotated ShipRSImageNet contains 17,573 ship instances. There are five critical contributions of the proposed ShipRSImageNet dataset compared with other existing remote sensing image datasets. Images are collected from various remote sensors cover- ing multiple ports worldwide and have large variations in size, spatial resolution, image quality, orientation, and environment. Ships are hierarchically classified into four levels and 50 ship categories. The number of images, ship instances, and ship cate- gories is larger than that in
This dataset comprises fractured and non-fractured X-ray images covering all anatomical body regions, including lower limb, upper limb, lumbar, hips, knees, etc. The dataset is categorized into train, test, and validation folders, each containing fractured and non-fractured radiographic images.
This dataset consists of both fractured and non-fractured X-ray images encompassing various anatomical regions of the body, such as the lower limb, upper limb, lumbar region, hips, knees, and more. It is organized into three main folders: train, test, and validation, each containing both fractured and non-fractured radiographic images. You can freely access the dataset via the following link: https://www.kaggle.com/datasets/bmadushanirodrigo/fracture-multi-region-x-ray-data/data
Consists of 36,785 images belonging to a diverse 92 classes. This class count is significantly higher than publicly available datasets. Maintains a low-class imbalance and a highly comprehensive data distribution for robust model training. It also provides the remote sensing community with an extra platform to validate the performance on multiple benchmarks.
a large video dataset captured with UAVs in different complex real-world scenes, with multiple representations, suitable for multi-task learning.
Dataset Description: Summarized Wiki Articles with TTL Knowledge Graphs
We designed an emotional speech database that can be used for emotion recognition as well as recognition and synthesis of speech with various emotions. The database was designed by compiling tweets acquired from Twitter and selecting emotion- dependent tweets considering phonetic and prosodic balance. We classified gathered tweets into four emotions: joy, anger, sadness and neutral, and then selected 50 sentences from sentences of each emotion based on the entropy-based algorithm. We compared the selected sentence sets with randomly selected sentence sets from aspects of phonetic and prosodic balance and sentence length, and confirmed that the sets selected by the algorithm were more balanced. Next, we recorded emotional speech based on the selected sentences. Then, we evaluated the speech from the viewpoint of emotional recognition and emotional speech recognition.
The Needle in a Needlestack (NIAN) is a new benchmark designed to measure how well Language Learning Models (LLMs) pay attention to the information in their context window¹.
The Remote Sensing dataset contains the following key features for each annotated marking:
The FlareReal600 is a nighttime flare removal dataset, which contains 650 real-captured images pairs and 500 flare images. The training set contains 600 images pairs and 500 flare images and the validation set contains 50 image pairs. Images pairs within the dataset are captured from various place (e.g., street, park, indoor) and under incorrect & correct exposure settings. Each flare-corrupted image contains various light sources. Flare images are captured from a dark room with multiple-color light sources.