Source-linked AI model and dataset records used to investigate visual perception tasks in UAV and robotics workflows.
25 evidence-linked resources in this topic
Building aerial visual perception systems: datasets, detection and segmentation models, tracking, and the edge deployment path.
Topic reviewed 2026-08-16 · Membership is an editorial index, not a compatibility or performance claim.
Stack layers: perception_localization · data · edge_ai · sensors
This topic is intentionally narrow: it connects the current release's AI model and dataset records, then explains how they should be read together without implying that a dataset proves model accuracy or that a model is deployable on a particular aircraft. OpenFly records the resource type, source tier, verification status, repository or project URL, and retrieval date separately. The topic page therefore supports an evidence review workflow: identify whether the item is a model or dataset, open the relevant SourceRef, inspect the recorded limitations, and use the linked guide or collection for deployment questions. Hardware acceleration, benchmark scores, latency, and field performance remain Unknown unless the release contains explicit evidence for them.
Topic membership is an editorial index, not a compatibility or performance claim. Verify each resource through its SourceRef, retrieval date, and verification status.
Aeroscapes: Aerial Semantic Segmentation Dataset Unofficial redistribution of the AeroScapes dataset under the original CC BY-SA 4.0 license. Disclaimer This repository is not an official release of the AeroScapes dataset. The AeroScapes dataset was created by Ishan Nigam, Chen Huang, and Deva Ramanan, with dataset collection and manual annotation supported by Autel Robotics. They retain all copyr
[ECCV 2022] ByteTrack: Multi-Object Tracking by Associating Every Detection Box
End-to-End Object Detection with Transformers
Drone-based Joint Density Map Estimation, Localization and Tracking with Space-Time Multi-Scale Attention Network
Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning
[ECCV 2024] Official implementation of the paper "Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection"
GWHD 2021: Global Wheat Head Dataset (Object Detection) Unofficial redistribution of the Global Wheat Head Dataset (GWHD) 2021 competition release, reformatted into a standardized YOLO-compatible directory layout. Disclaimer This repository is not an official release of the Global Wheat Head Dataset. GWHD was created by the Global Wheat Head Detection consortium — a multi-institution, multi-countr
A high-altitude infrared thermal dataset for Unmanned Aerial Vehicle-based object detection
Inria Aerial Image Labeling Dataset Description The Inria Aerial Image Labeling Dataset is a building semantic segmentation dataset proposed in "Can semantic labeling methods generalize to any city? the inria aerial image labeling benchmark," Maggiori et al.. It consists of 360 high-resolution (0.3m) RGB images, each with a size of 5000x5000 pixels. These images are extracted from various internat
Python implementation of the IOU Tracker
Example and utiliy scripts for the Mid-Air dataset
OpenMMLab Detection Toolbox and Benchmark
OpenMMLab Rotated Object Detection Toolbox and Benchmark
OpenMMLab Semantic Segmentation Toolbox and Benchmark.
SeaDronesSee: Maritime UAV Object Detection Dataset Unofficial redistribution of the SeaDronesSee object-detection (v2) dataset, reformatted into a standardized YOLO-compatible directory layout. Disclaimer This repository is not an official release of the SeaDronesSee dataset. SeaDronesSee was created by Leon Amadeus Varga, Benjamin Kiefer, Martin Messmer, and Andreas Zell at the University of Tüb
UAVid: Aerial Semantic Segmentation Dataset Unofficial redistribution of the UAVid dataset under the original CC BY-NC-SA 4.0 license. Disclaimer This repository is not an official release of the UAVid dataset. The UAVid dataset was created by the original authors, who retain all copyright and intellectual property rights. This repository does not claim ownership of any images, annotations, or met
Ultralytics YOLO26, YOLO11, YOLOv8 — object detection, instance segmentation, semantic segmentation, image classification, pose estimation, object tracking
This dataset was created using LeRobot. UZH-FPV Drone Racing, 128px, LeRobot format A processed derivative of the UZH-FPV Drone Racing Dataset packaged as a LeRobotDataset (v3): 24 episodes, 24,242 frames of aggressive indoor FPV racing flight, as 128x128 grayscale onboard camera frames with 4D continuous control actions. This is not my recording. All flight data was collected and published by the
The dataset for drone based detection and tracking is released, including both image/video, and annotations.
VisDrone aerial object detection toolkit with 33 models (Torchvision + YOLO), training, evaluation, video inference, benchmarking, and annotation conversion.
Official code for VoLN: Vision-Only Long-Horizon Navigation—Paradigm, Benchmark, and Method. An embodied AI UAV benchmark and VoLN-MLLM agent bridging VLN and multimodal LLMs across 7,210 episodes, AirSim, and real-world flights.
Ultralytics YOLOv5 in PyTorch for object detection, instance segmentation, classification, training, and export.
YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported. Documentation: https://yolox.readthedocs.io/