FastestDetV2 [中文]

🔥🔥Even faster and stronger than FastestDet🔥🔥
🔥🔥比FastestDet更快更强🔥🔥

Improvements

• ⚡2.5% mAP50 & 1% mAP50:95 improvement, with ~20% faster speed compared to FastestDet
• Assign Guidance Module and SimOTA label assignment for better precision
• Quantization-aware, reparameterizable MobileOne backbone and convolution modules
• ⚡相比FastestDet，mAP50提升2.5%，mAP50:95提升1%，同时速度提升约20%
• 采用Assign Guidance Module和SimOTA标签分配策略，以获得更好的精度
• 基于支持量化感知训练、可重参数化的MobileOne骨干网络和卷积模块

Gallery

Benchmarks

Model	mAP50	mAP50:95	Resolution	Inference time (4x core)	Inference time (1x core)	Params (M)
FastestDetV2	27.8%	14.0%	352X352	2.83ms	6.95ms	0.33M
FastestDetV2-2x	36.6%	19.9%	352X352	6.81ms	19.88ms	1.22M
FastestDet	25.3%	13.0%	352X352	3.68ms	8.48ms	0.24M
NanoDet-m	-	20.6%	320X320	7.76ms	22.23ms	0.95M
YOLOX-Nano	-	25.8%	416X416	36.88ms	92.52ms	0.91M
YOLOv8n	56.8%	37.4%	640X640	57.03ms	122.63ms	7.2M

Tested on EmbedFire LubanCat-4 RK3588S ARM 4*Cortex-A76 CPU@2.0GHz, using NCNN.

Multi-platform Benchmarks

Device	Computing backend	System	Framework	Inference time (4x core)	Inference time (1x core)	2x Inference time (4x core)	2x Inference time (1x core)
EmbedFire LubanCat-4	RK3588 (CPU) 1	Linux (arm64)	NCNN	2.83ms	6.95ms	6.81ms	19.88ms
EmbedFire LubanCat-4	RK3588 (NPU)	Linux (arm64)	RKNN	7.067ms 2	7.532ms	8.04ms 3	9.56ms
Google Pixel 10 Pro XL	Tensor G5 (CPU)	Android (arm64)	NCNN	2.69ms	3.88ms	4.66ms	6.26ms
OnePlus 6	Snapdragon 845 (CPU)	Android (arm64)	NCNN	4.73ms	8.14ms	11.56ms	17.84ms
Dell Precision 3630 Tower	Core i9-9900 (CPU) 4	Linux (x86_64)	NCNN	2.90m	7.31ms	6.86ms	19.94ms

¹: At 2.0 GHz.
², ³: RKNNLite.NPU_CORE_0_1_2 is used.
⁴: At 800MHz.

Model Zoo

Download	Note
fastestdetv2.pth, fastestdetv2_unfused.pth fastestdetv2-2x.pth, fastestdetv2-2x_unfused.pth	Model weights
qamobileone.pth qamobileone-2x.pth	Backbone weights
fastestdetv2.apk	Android demo
fastestdetv2.bin, fastestdetv2.param fastestdetv2-2x.bin, fastestdetv2-2x.param	NCNN files
fastestdetv2.rknn fastestdetv2-2x.rknn	RKNN files
fastestdetv2.onnx fastestdetv2-2x.onnx	ONNX files
fastestdetv2.pt, fastestdetv2_ptq.arm.pt, fastestdetv2_ptq.x86.pt fastestdetv2-2x.pt, fastestdetv2-2x_ptq.arm.pt, fastestdetv2-2x_ptq.x86.pt	TorchScript files

Usage

Dependencies

pip install -r requirements.txt

Datasets & Configurations

Datasets can be either in Darknet format (like FastestDet, using a text file to list image paths, with labels stored in separate .txt files in the same directory) or in YOLO format (like YOLOv8, where each image has a corresponding .txt label file in a seperate directory). Labels are in cls cx cy w h normalized bboxes.

The .yaml configurations file specifies dataset paths, model settings, and training hyperparameters. Dataset could be either in Darknet format or YOLO format. Class names can also be in a single text file with each line representing a class name. See configs/coco.yaml for example.

Evaluation & Testing

You can evaluate the model with a fused (reparameterized) model weights file.

python3 eval.py --configs CONFIGS_PATH --weight WEIGHTS_PATH

Or test it on an image:

python3 test.py --configs CONFIGS_PATH --weights WEIGHTS_PATH --image IMAGE_PATH

Training

Download the backbone weights and place it under weights/qamobileone.pth and weights/qamobileone-2x.pth, and run:

python3 train.py --configs CONFIGS_PATH

Or finetune it with an unfused weights file:

python3 train.py --configs CONFIGS_PATH --weights WEIGHTS_PATH

Deployment

ONNX & TorchScript

Export to ONNX and TorchScript format with:

python3 test.py --configs CONFIGS_PATH --weights WEIGHTS_PATH --export

PT2E PTQ

Post-training quantization for x86 (with X86InductorQuantizer) or arm (with XNNPackQuantizer) platforms, with fused weights:

python3 quant.py --configs CONFIGS_PATH --weights WEIGHTS_PATH --image IMAGE_PATH --target TARGET_PLATFORM

NCNN

Follow deploy/ncnn/README.md or deploy/ncnn_android/README.md (for Android).

RKNN

Follow deploy/rknn/README.md.

Citation

@misc{=FastestDetV2,
    title={FastestDetV2: Even faster and stronger than FastestDet},
    author={Pairman},
    howpublished = {\url{https://github.com/Pairman/FastestDetV2}},
    year={2025}
}

References

• FastestDet: https://github.com/dog-qiuqiu/FastestDet
• Assign Guidance Module and NCNN deployment: https://github.com/RangiLyu/nanodet
• MobileOne: https://github.com/apple/ml-mobileone and https://github.com/glory-wan/TF-Net
• Quantization-aware RepConv: https://github.com/meituan/YOLOv6
• SimOTA label assignment: https://github.com/open-mmlab/mmyolo
• NCNN: https://github.com/Tencent/ncnn