A collection of YOLO object detection models fine-tuned on the KITTI benchmark for autonomous-driving street-scene detection.
š¤ Open to Collab
Saumya Saksena
dronefreak
AI & ML interests
Computer Vision, Deep Learning, Image Restoration, Image De-noising, LLMs, RAG, Image Classification, Image Segmentation, EEG Classification, Signal Processing, PPG, BCI
Recent Activity
posted an update about 14 hours ago
š Open-sourcing the KITTI Object Detection Model Zoo on Hugging Face.
- š¤ 12 models: YOLOv8, YOLOv9, YOLO11 and YOLO26, from nano/tiny up to x-large.
- š Street-scene detection: cars, cyclists, pedestrians, vans, trucks and more, in KITTI's ultra-wide frames.
- š Model cards with metrics, per-class results, curves, showcases, a demo video and full configs.
Headline numbers:
- š Best mAP@50: 43.73% (YOLO26x). Best mAP@50:95: 26.54% (YOLO26s).
- ā” YOLO26n gets 42.54% mAP@50 at just 6.1 GFLOPs, within 1.2 points of YOLO26x at ~34x fewer FLOPs.
Trained and evaluated with DetectionBench: https://github.com/dronefreak/DetectionBench
Dataset credit: Andreas Geiger, Philip Lenz and Raquel Urtasun (CVPR 2012). This is an unofficial YOLO-ready reformat (CC BY-NC-SA 3.0). Metrics are on the validation split, since KITTI has no public test labels.
š¦ Dataset: dronefreak/KITTI
š¤ Collection: https://huggingface.co/collections/dronefreak/kitti-object-detection-model-zoo-6ab4cb94fb08dccf3bde2e8b
Feedback and contributions welcome. updated a collection 2 days ago
KITTI Object Detection Model Zoo updated a model 2 days ago
dronefreak/kitti-yolov9tOrganizations
A collection of YOLO and RF-DETR object detection models fine-tuned on the HRP4K benchmark for high-resolution road pothole detection.
A collection of YOLO object detection models fine-tuned on the BDD100K benchmark for autonomous-driving scene detection. RF-DETR variants to follow.
Unofficial attributed mirrors of the non-ViT Co-DINO COCO checkpoints (5-scale R50 / Swin-L, ICCV 2023), previously Drive-only. license: unknown.
-
dronefreak/co-dino-5scale-r50-1x-coco
Object Detection ā¢ Updated ā¢ 35 -
dronefreak/co-dino-5scale-lsj-r50-1x-coco
Object Detection ā¢ Updated ā¢ 32 -
dronefreak/co-dino-5scale-lsj-r50-3x-coco
Object Detection ā¢ Updated ā¢ 26 -
dronefreak/co-dino-5scale-9encoder-lsj-r50-1x-coco
Object Detection ā¢ Updated ā¢ 37
A small collection of end-to-end adverse weather effects removal AI models from single images
A collection of YOLO and RF-DETR object detection models fine-tuned on the SeaDronesSee benchmark for maritime search-and-rescue detection from UAVs.
A collection of YOLO and RF-DETR object detection models fine-tuned on the ExDark (Exclusively Dark) benchmark for detection in low-light images.
A collection of CABiNet and YOLO26 models trained and evaluated under a common pipeline for aerial semantic segmentation on VDD 2025 dataset.
A collection of CABiNet and YOLO26 models trained and evaluated under a common pipeline for aerial semantic segmentation.
A collection of YOLO and RF-DETR object detection models fine-tuned on the PKLot benchmark for parking-space occupancy detection.
A collection of YOLO and RF-DETR object detection models fine-tuned on the RDD2022 benchmark for road damage detection (cracks and potholes).
A collection of YOLO and RF-DETR object detection models fine-tuned on the UAVDT benchmark for vehicle detection in aerial drone footage.
Unofficial attributed mirrors of the non-ViT Co-Deformable-DETR COCO checkpoints (ICCV 2023), previously Google-Drive-only. license: unknown.
-
dronefreak/co-deformable-detr-r50-1x-coco
Object Detection ā¢ Updated ā¢ 65 -
dronefreak/co-deformable-detr-swin-t-1x-coco
Object Detection ā¢ Updated ā¢ 33 -
dronefreak/co-deformable-detr-swin-t-3x-coco
Object Detection ā¢ Updated ā¢ 33 -
dronefreak/co-deformable-detr-swin-s-1x-coco
Object Detection ā¢ Updated ā¢ 32
A collection of YOLO object detection models fine-tuned on the Brackish benchmark for underwater marine animal detection. RF-DETR variants to follow.
A collection of YOLO and RF-DETR object detection models fine-tuned on the Global Wheat Head Dataset (GWHD) for wheat head detection in field images.
A collection of CABiNet and YOLO26 models trained and evaluated under a common pipeline for aerial semantic segmentation on Aeroscapes dataset.
-
dronefreak/aeroscapes-yolo26m-sem
Image Segmentation ā¢ Updated ā¢ 17 ā¢ 3 -
dronefreak/aeroscapes-yolo26x-sem
Image Segmentation ā¢ Updated ā¢ 23 ā¢ 3 -
dronefreak/aeroscapes-cabinet-mobilenetv3-large
Image Segmentation ā¢ Updated ā¢ 7 ā¢ 2 -
dronefreak/aeroscapes-yolo26l-sem
Image Segmentation ā¢ Updated ā¢ 17 ā¢ 2
A collection of YOLO and RF-DETR object detection models fine-tuned on the LISA Traffic Lights dataset for traffic light detection in driving scenes.
A collection of YOLO and RF-DETR object detection models fine-tuned on the VisDrone benchmark for object detection in drone imagery.
A collection of YOLO object detection models fine-tuned on the KITTI benchmark for autonomous-driving street-scene detection.
A collection of YOLO and RF-DETR object detection models fine-tuned on the PKLot benchmark for parking-space occupancy detection.
A collection of YOLO and RF-DETR object detection models fine-tuned on the HRP4K benchmark for high-resolution road pothole detection.
A collection of YOLO and RF-DETR object detection models fine-tuned on the RDD2022 benchmark for road damage detection (cracks and potholes).
A collection of YOLO object detection models fine-tuned on the BDD100K benchmark for autonomous-driving scene detection. RF-DETR variants to follow.
A collection of YOLO and RF-DETR object detection models fine-tuned on the UAVDT benchmark for vehicle detection in aerial drone footage.
Unofficial attributed mirrors of the non-ViT Co-DINO COCO checkpoints (5-scale R50 / Swin-L, ICCV 2023), previously Drive-only. license: unknown.
-
dronefreak/co-dino-5scale-r50-1x-coco
Object Detection ā¢ Updated ā¢ 35 -
dronefreak/co-dino-5scale-lsj-r50-1x-coco
Object Detection ā¢ Updated ā¢ 32 -
dronefreak/co-dino-5scale-lsj-r50-3x-coco
Object Detection ā¢ Updated ā¢ 26 -
dronefreak/co-dino-5scale-9encoder-lsj-r50-1x-coco
Object Detection ā¢ Updated ā¢ 37
Unofficial attributed mirrors of the non-ViT Co-Deformable-DETR COCO checkpoints (ICCV 2023), previously Google-Drive-only. license: unknown.
-
dronefreak/co-deformable-detr-r50-1x-coco
Object Detection ā¢ Updated ā¢ 65 -
dronefreak/co-deformable-detr-swin-t-1x-coco
Object Detection ā¢ Updated ā¢ 33 -
dronefreak/co-deformable-detr-swin-t-3x-coco
Object Detection ā¢ Updated ā¢ 33 -
dronefreak/co-deformable-detr-swin-s-1x-coco
Object Detection ā¢ Updated ā¢ 32
A small collection of end-to-end adverse weather effects removal AI models from single images
A collection of YOLO object detection models fine-tuned on the Brackish benchmark for underwater marine animal detection. RF-DETR variants to follow.
A collection of YOLO and RF-DETR object detection models fine-tuned on the SeaDronesSee benchmark for maritime search-and-rescue detection from UAVs.
A collection of YOLO and RF-DETR object detection models fine-tuned on the Global Wheat Head Dataset (GWHD) for wheat head detection in field images.
A collection of YOLO and RF-DETR object detection models fine-tuned on the ExDark (Exclusively Dark) benchmark for detection in low-light images.
A collection of CABiNet and YOLO26 models trained and evaluated under a common pipeline for aerial semantic segmentation on Aeroscapes dataset.
-
dronefreak/aeroscapes-yolo26m-sem
Image Segmentation ā¢ Updated ā¢ 17 ā¢ 3 -
dronefreak/aeroscapes-yolo26x-sem
Image Segmentation ā¢ Updated ā¢ 23 ā¢ 3 -
dronefreak/aeroscapes-cabinet-mobilenetv3-large
Image Segmentation ā¢ Updated ā¢ 7 ā¢ 2 -
dronefreak/aeroscapes-yolo26l-sem
Image Segmentation ā¢ Updated ā¢ 17 ā¢ 2
A collection of CABiNet and YOLO26 models trained and evaluated under a common pipeline for aerial semantic segmentation on VDD 2025 dataset.
A collection of YOLO and RF-DETR object detection models fine-tuned on the LISA Traffic Lights dataset for traffic light detection in driving scenes.
A collection of CABiNet and YOLO26 models trained and evaluated under a common pipeline for aerial semantic segmentation.
A collection of YOLO and RF-DETR object detection models fine-tuned on the VisDrone benchmark for object detection in drone imagery.