Papers › AQD: Towards Accurate Fully-Quantized Object Detection

AQD: Towards Accurate Fully-Quantized Object Detection

14 Jul 2020CVPR 2021 1arXiv:2007.06919archive 2025-07-28

Peng Chen, Jing Liu, Bohan Zhuang, Mingkui Tan, Chunhua Shen

Network quantization allows inference to be conducted using low-precision arithmetic for improved inference efficiency of deep neural networks on edge devices. However, designing aggressively low-bit (e.g., 2-bit) quantization schemes on complex tasks, such as object detection, still remains challenging in terms of severe performance degradation and unverifiable efficiency on common hardware. In this paper, we propose an Accurate Quantized object Detection solution, termed AQD, to fully get rid of floating-point computation. To this end, we target using fixed-point operations in all kinds of layers, including the convolutional layers, normalization layers, and skip connections, allowing the inference to be executed using integer-only arithmetic. To demonstrate the improved latency-vs-accuracy trade-off, we apply the proposed methods on RetinaNet and FCOS. In particular, experimental results on MS-COCO dataset show that our AQD achieves comparable or even better performance compared with the full-precision counterpart under extremely low-bit schemes, which is of great practical value. Source code and models are available at: https://github.com/ziplab/QTool

PaperPDFConference PDFCode

Code

ziplab/qtool officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image ClassificationObjectObject DetectionQuantizationobject-detection

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

1x1 ConvolutionBatch NormalizationConvolutionFCOSFPNFocal LossNon Maximum SuppressionRetinaNet

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections