Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
16,462 datasets
ImageNet-REPA-Latents is a dataset hosted on Kaggle. The title suggests it contains latent representations, likely derived from the ImageNet image database. The dataset's specific content, scale, and creation details require verification after download.
dhbk_hb_model cnn gei 060 v24.7 is a pre-trained model dataset hosted on Kaggle. The title suggests it is a Convolutional Neural Network (CNN) model, likely version 24.7. Its specific architecture, training data, and performance metrics are not detailed in the provided metadata.
An implementation of the EfficientNet architecture for TensorFlow Keras. It is tagged for use in Image Classification and Computer Vision tasks. No information is available on the number of rows, columns, or specific data content.
A dataset likely containing images of source code for optical character recognition tasks, sourced from Kaggle. The dataset's specific size, creation date, and author are unknown from the provided metadata. Its content and structure require verification after download.
5,231 high-quality AI-generated images form this curated collection developed by ArkDevelopmentLabs / ArkAiLab (ADL). The Nexora Vision Dataset v1 Medium is an expanded release built for research, training, and evaluation of generative AI systems. It contains cinematic, realistic, anime, sci-fi, and artistic images, with the dataset last updated on December 15, 2025.
VINDR YOLO 1024 is a Kaggle-hosted dataset with a title suggesting a focus on medical image analysis. The dataset's name implies it is formatted for the YOLO object detection framework, likely at a 1024x1024 resolution. Further details on its size, origin, and specific annotations are not provided in the available metadata.
ResnetLite is a pre-trained model dataset hosted on Kaggle. The dataset likely contains model weights and architecture files for a lightweight ResNet variant. Its specific size, author, and last update date are unknown.
Mask R-CNN Weights is a pre-trained model for computer vision tasks, hosted on Kaggle. The dataset likely contains model weights for the Mask R-CNN architecture, a neural network designed for instance segmentation. Specific details on the training data, model version, and performance metrics are not provided in the available metadata.
MMDoc is a dataset containing real-world document images with OCR results and translations across 10 language pairs. It is designed for training and evaluating multimodal document translation models that understand visual layout and context. The dataset was created by rileykim and was last updated on 2025-12-22.
Model weights for a Convolutional Neural Network trained on the MNIST dataset. The platform tags suggest this dataset contains the learned parameters from a deep learning model for handwritten digit recognition. Published on Kaggle, its specific architecture, training details, and performance metrics are unknown from the provided metadata.
Kaggle hosts a dataset titled 'cnn_dailymail_preprocessed_edu_spacy', which likely contains news articles from CNN and the Daily Mail. The dataset appears to be preprocessed, potentially for educational purposes or using the spaCy NLP library. Its specific scale, authorship, and update history are not provided in the available metadata.
A pre-trained Convolutional Neural Network model, likely for image classification or feature extraction, published on Kaggle. The specific architecture, training data, and performance metrics are not detailed in the available metadata. Users must download the model to verify its intended application and capabilities.
The dataset 'dhbk_hb_emb cnn gei 270 v24.14' is hosted on Kaggle. Its title and platform tags suggest it contains data for computer vision tasks, likely involving Convolutional Neural Network (CNN) embeddings. The specific content, size, and authorship details are not provided in the available metadata.
A dataset from Kaggle's Research category focused on computational docking simulations. The data likely contains results from hybrid quantum-chemical methods used to guide the adaptive docking of macrocyclic ligands into a target. The specific author, organization, and data volume are unknown.
WLASL-126Keypoints-2000 likely contains pose estimation data for sign language gestures. The dataset appears to be derived from the WLASL (World-Level American Sign Language) dataset, focusing on 126 keypoints across 2000 instances. It is hosted on Kaggle, but the author, organization, and specific collection details are unknown.
Global Wheat YOLO11s Best Weights likely contains pre-trained model weights for a YOLO-based object detection system. The dataset is published on Kaggle, but its author, organization, and creation date are unknown. Columns suggest it may include weights files and possibly associated metadata.
PCB Defect Detection Merged Dataset 2026 YOLOv8 v1 is a dataset hosted on Kaggle. Its title suggests it contains images of printed circuit boards annotated for defect detection tasks. The dataset is likely intended for training object detection models, specifically the YOLOv8 architecture.
A pre-trained machine learning model for bird species classification from audio data, likely derived from the BirdCLEF 2026 competition. The model is provided in the ONNX format, which suggests it is optimized for efficient inference across different frameworks. It is hosted on Kaggle under the 'Pre Trained Model' tag.
A dataset published on Kaggle for object detection tasks, likely containing images. The dataset name and platform tags suggest it is designed for use with YOLO (You Only Look Once) models. Specific details on size, creator, and update date are not provided in the metadata.
A dataset from Kaggle containing images of IEC electrical symbols, likely formatted for training YOLO v6 object detection models. The dataset's specific size, creation date, and author are unknown. Its content is inferred from the title and platform tags.