real-time object detection
Official PyTorch implementation of YOLOv7, a real-time single-stage object detection model with training and inference code.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
Choose for a current, actively maintained all-in-one YOLO framework with YOLOv8/11 and segmentation/pose/classification support.
Ultralytics YOLO26, YOLO11, YOLOv8 — object detection, instance segmentation, semantic segmentation, image classification, pose estimation, object tracking
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
Choose when you prefer a maintained PaddlePaddle framework and need many detector/segmentation/tracking models rather than one YOLOv7 PyTorch repo.
Object Detection toolkit based on PaddlePaddle. It supports object detection, instance segmentation, multiple object tracking and real-time multi-person keypoint detection.
Choose when you need real-time open-vocabulary detection from arbitrary text prompts, not YOLOv7's fixed COCO-style classes.
[CVPR 2024] Real-Time Open-Vocabulary Object Detection
Choose if you want a config-driven toolbox of many detector architectures; note it has had no pushes for two years.
OpenMMLab Detection Toolbox and Benchmark
Choose for the original transformer end-to-end DETR baseline (Apache-2.0) if you accept an archived repo, not YOLOv7 real-time speed.
End-to-End Object Detection with Transformers
Choose if you want a maintained real-time DETR-style detector with detection+segmentation and Apache-2.0 code.
RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning. [ICLR 2026]
Choose for an MIT-licensed center-point/heatmap detector if you want no anchor priors, accepting it is much older/unmaintained.
Object detection, 3D detection, and pose estimation using center point detection:
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
Choose only for legacy YOLOv6-style deployments; it is in the same YOLO category but has not been maintained for over two years.
YOLOv6: a single-stage object detection framework dedicated to industrial applications.
Choose for an older MIT-licensed YOLOv4 PyTorch implementation when you specifically want YOLOv4-style code, not YOLOv7.
这是一个YoloV4-pytorch的源码，可以用于训练自己的模型。
Choose for an older mobile/compression-focused YOLO-style detector in the YOLObile research code if you target on-device experiments.
This is the implementation of YOLObile: Real-Time Object Detection on Mobile Devices via Compression-Compilation Co-Design
Choose for a Keras-only YOLOv4 workflow; it is a previous-generation YOLO implementation and not maintained.
这是一个YoloV4-keras的源码，可以用于训练自己的模型。
These projects were analysed and named yolov7 among their alternatives. The relationship is not symmetric — how yolov7 rates them is a separate judgement, made when yolov7 is analysed in its own right.
calls yolov7 “Choose when you want the YOLOv7 architecture and can migrate from the YOLOv3 codebase.”
PyTorch implementation of YOLOv3, YOLOv3-SPP, and YOLOv3-tiny for real-time object detection with training, validation, inference, and multi-format export.
calls yolov7 “Choose when you want a real-time GPL-3.0 YOLOv7 model with proven deployment, not the center-point/pose/3D model.”
Object detection, 3D detection, and pose estimation using center point detection:
calls yolov7 “For reproducing the YOLOv7 recipe; unmaintained for two years and requires fixed classes.”
[CVPR 2024] Real-Time Open-Vocabulary Object Detection