# rangilyu/nanodet

**Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/rangilyu-nanodet).**

6,222 stars · 1,105 forks · Python · Apache-2.0

## Links

- GitHub: https://github.com/RangiLyu/nanodet
- awesome-repositories: https://awesome-repositories.com/repository/rangilyu-nanodet.md

## Topics

`anchor-free` `android` `deep-learning` `deep-neural-networks` `efficientnet` `mnn` `model-zoo` `nanodet` `nanodet-plus` `ncnn` `object-detection` `openvino` `pytorch` `repvgg` `shufflenet`

## Description

NanoDet-Plus⚡Super fast and lightweight anchor-free object detection model. 🔥Only 980 KB(int8) / 1.8MB (fp16) and run 97FPS on cellphone🔥

## Tags

### Artificial Intelligence & ML

- [Lightweight Anchor-Free Detectors](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-detection-tracking/edge-object-detection/lightweight-anchor-free-detectors.md) — Ships a 980KB anchor-free detection model achieving 97 FPS on mobile devices for real-time edge inference.
- [Lightweight Anchor-Free Detections](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-detection-tracking/object-detection/anchor-free-detection-models/lightweight-anchor-free-detections.md) — Provides a lightweight anchor-free detection model optimized for speed and accuracy on resource-constrained devices.
- [Anchor-Free Detection Logic](https://awesome-repositories.com/f/artificial-intelligence-ml/anchor-box-systems/anchor-free-detection-logic.md) — Implements anchor-free detection logic that predicts bounding boxes directly from feature map points.
- [GhostNet Backbones](https://awesome-repositories.com/f/artificial-intelligence-ml/backbone-integrations/ghostnet-backbones.md) — Provides a GhostNet backbone that reduces parameters and FLOPs through cheap linear operations.
- [Real-Time Object Detection](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-detection-tracking/real-time-object-detection.md) — Achieves 97 FPS real-time object detection on mobile devices for live video stream analysis.
- [Depthwise Separable Convolutions](https://awesome-repositories.com/f/artificial-intelligence-ml/convolutional-neural-networks/pointwise-convolutions/depthwise-separable-convolutions.md) — Uses depthwise separable convolutions to shrink model size and accelerate inference.
- [Detection Model Training](https://awesome-repositories.com/f/artificial-intelligence-ml/custom-model-training/detection-model-training.md) — Provides a training pipeline for custom object detection models with configurable backbones and data augmentation. ([source](https://cdn.jsdelivr.net/gh/rangilyu/nanodet@main/README.md))
- [Light-Weight Detection Necks](https://awesome-repositories.com/f/artificial-intelligence-ml/inference-scaling/resolution-scaling/hierarchical-feature-pyramids/multi-scale-feature-pyramids/multi-scale-feature-aggregation/light-weight-detection-necks.md) — Ships a light-weight detection neck that aggregates multi-scale features with minimal overhead.
- [Edge AI Model Deployment](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/model-deployment-and-serving/local-and-on-device-inference/edge-ai-model-deployment.md) — Exports and deploys detection models to edge hardware using ONNX, ncnn, MNN, or OpenVINO.
- [ONNX Model Exporters](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/infrastructure/model-deployment-and-serving/serialization-and-export-formats/onnx-model-exporters.md) — Exports object detection models to ONNX format for cross-platform deployment.
- [Mobile-Optimized Neural Networks](https://awesome-repositories.com/f/artificial-intelligence-ml/neural-network-architectures/mobile-optimized-neural-networks.md) — Designs neural network architectures for high-speed object detection on ARM CPUs and mobile platforms.
- [PyTorch-to-ONNX Pipelines](https://awesome-repositories.com/f/artificial-intelligence-ml/pytorch-model-export/pytorch-to-onnx-pipelines.md) — Converts trained PyTorch models to ONNX format for deployment across multiple inference engines.
- [Python API](https://awesome-repositories.com/f/artificial-intelligence-ml/training-configurations/python-api.md) — Provides a Python API for running object detection inference on images, videos, or webcam streams. ([source](https://cdn.jsdelivr.net/gh/rangilyu/nanodet@main/README.md))
- [ARM-Optimized Detectors](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-detection-tracking/edge-object-detection/local-object-detection/arm-optimized-detectors.md) — Delivers a 1.8MB model optimized for ARM CPUs achieving 97 FPS on mobile devices. ([source](https://cdn.jsdelivr.net/gh/rangilyu/nanodet@main/README.md))
- [Multi-Backend Inference Support](https://awesome-repositories.com/f/artificial-intelligence-ml/inference-backends/multi-backend-inference-support.md) — Supports running inference across multiple backends including ncnn, MNN, OpenVINO, and web browsers. ([source](https://cdn.jsdelivr.net/gh/rangilyu/nanodet@main/README.md))
- [Assign Guidance Modules](https://awesome-repositories.com/f/artificial-intelligence-ml/label-assignment-strategies/assign-guidance-modules.md) — Ships an Assign Guidance Module that improves training target assignment using classification and regression quality.
- [Dynamic Soft Label Assignments](https://awesome-repositories.com/f/artificial-intelligence-ml/label-assignment-strategies/dynamic-soft-label-assignments.md) — Implements dynamic soft label assignment to balance positive and negative samples during training.

### Graphics & Multimedia

- [Real-Time Model Inference on Frames](https://awesome-repositories.com/f/graphics-multimedia/video-frame-processing/real-time-model-inference-on-frames.md) — Performs object detection on live video streams at high frame rates on low-power hardware.
- [Edge Deployments](https://awesome-repositories.com/f/graphics-multimedia/video-frame-processing/real-time-model-inference-on-frames/edge-deployments.md) — Deploys trained detection models to edge devices for real-time inference using ONNX and ncnn/MNN/OpenVINO. ([source](https://cdn.jsdelivr.net/gh/rangilyu/nanodet@main/README.md))