Command Palette
Search for a command to run...
Dataset Summary | Essential Datasets for Beginners to Learn YOLO, Covering Animals, Crops, Facial Expressions, and More
For those who are new to artificial intelligence and computer vision, the first "small goal" they set for themselves is often to run a YOLO model. When I was learning, I was still exposed to YOLOv8, and the first project I did was "fruit detection based on YOLO".The YOLO series has been upgraded to YOLOv13, which has improved both performance and efficiency.
The latest YOLOv13 introduces the Hypergraph-based Adaptive Correlation Enhancement (HyperACE) mechanism, which significantly improves detection performance in complex scenarios.At the same time, the FullPAD paradigm is adopted to inject the enhanced feature flow into the entire process of Backbone, Neck and Head, achieving stronger information collaboration.Based on this, YOLOv13-N improves mAP by 3.0% compared to YOLO11-N and by 1.5% compared to YOLOv12-N. YOLOv13's breakthrough in real-time object detection also provides new ideas for the design of subsequent efficient and powerful detection systems.
Thanks to the contributions of the open source community, the YOLO model has seen rapid upgrades from generation to generation. However, as with so many great things, choosing the right dataset is crucial for beginners. HyperAI has compiled some datasets suitable for beginners to experiment with YOLO, hoping to help you easily get started.
Click to view more open source datasets:
Aerial photography dataset summary
1 , Bird Species bird classification image dataset
Estimated size:1.14 GB
Download address:https://go.hyper.ai/DvNwq
Bird Species is a bird image classification dataset suitable for training computer vision models to identify and classify bird species. The dataset contains 7 different species, each with 1,200 images. The images of each species contain the feather pattern, color, and body structure of the bird of that species. Some of the images are intentionally blurred, tilted, or contain 2 birds of different species, which increases the complexity of the real world and makes the model more robust for accurate classification in natural environments.
2 , Vegetable Identification Vegetable Image Recognition Dataset
Estimated size:2.26 GB
Download address:https://go.hyper.ai/UGczX
The dataset contains images of six types of vegetables: eggplant, beans, okra, squash, potatoes, and onions, with 800 images of each type, for a total of 4,800 images. It aims to enhance the capabilities of machine learning and computer vision in vegetable detection, classification, and recognition.
3 , Crops Disease Crop Disease Dataset
Estimated size:1.99 GB
Download address:https://go.hyper.ai/bD5vr
Crops Disease is an agricultural crop disease image dataset designed to help develop computer vision models to automatically detect and classify diseases of different crops. The dataset contains about 1,300 crop disease images, covering common diseases of a variety of crops (such as corn, tomatoes, potatoes, etc.), and each image is annotated with a specific disease category.
4 , Facial Expressions YOLO format detection dataset
Estimated size:3.77 GB
Download address:https://go.hyper.ai/ZaUdh
The dataset contains approximately 70,000 images, covering 9 facial expression categories, and taking into account both basic and complex emotions. It is suitable for application scenarios such as emotion recognition in computer vision, human-computer interaction, mental health analysis, and intelligent monitoring.
5 , Road Hazards Road Hazards Dataset
Estimated size:1 GB
Download address:https://go.hyper.ai/fEVuJ
The Road Hazards dataset contains 2.7k images and is primarily used to detect potholes, cracks, and open manholes on roads. This dataset has been augmented to increase data diversity and robustness. The images are divided into training and validation sets, and are classified into three different categories.
6 , Dog Breeds Dog breed image dataset
Download address:https://go.hyper.ai/M8rqJ
Dog Breeds is a dog breed image dataset that contains images of different dog breeds, designed to help train and evaluate dog breed classification models. The dataset contains thousands (over 17,000) images of different dog breeds, more than 100 breeds (terriers, hounds, mastiffs, spaniels, bichon frise, etc.), and is designed to help develop dog breed recognition systems.
7 , 15 animal image classification datasets
Estimated size:37.06 MB
Download address:https://go.hyper.ai/5SgsE
This dataset contains image folders of 15 animals: bear, bird, cat, cow, deer, dog, dolphin, elephant, giraffe, horse, kangaroo, lion, panda, tiger, and zebra. Each folder represents a class corresponding to the folder name and contains at least 120 images. All images are 224x224, suitable for image classification. The images were downloaded from the internet and preprocessed (resizing and augmentation) using the OpenCV library. Therefore, this dataset can be used directly for training without further data augmentation.
8 , UAVDT drone target detection and tracking video dataset
Estimated size:1.14 GB
Download address:https://go.hyper.ai/0f4Ej
UAVDT stands for "UAV Object Detection and Tracking Large-Scale Video Dataset." It contains 10 hours of raw video and approximately 8,000 representative video frames with manually annotated bounding boxes and useful labels such as vehicle category and occlusion. This dataset, captured by drones in various complex scenarios, primarily addresses three basic tasks: object detection (DET), single object tracking (SOT), and multiple object tracking (MOT).
The above is the aerial photography dataset compiled by HyperAI. If you have resources that you want to include on the hyper.ai official website, you are welcome to leave a message or submit your contribution to tell us!
About HyperAI
HyperAI (hyper.ai) is the leading artificial intelligence and high-performance computing community in China.We are committed to becoming the infrastructure in the field of data science in China and providing rich and high-quality public resources for domestic developers. So far, we have:
* Provide domestic accelerated download nodes for 1200+ public data sets
* Includes 300+ classic and popular online tutorials
* Interpretation of 200+ AI4Science paper cases
* Support 500+ related terms search
* Hosting the first complete Apache TVM Chinese documentation in China
Visit the official website to start your learning journey: