物件偵測中輸入退化對量化之魯棒性

摘要

後訓練量化（PTQ）對於在資源受限設備上部署高效能的物體檢測模型（如YOLO）至關重要。然而，降低精度對模型在面對現實世界輸入退化（如噪聲、模糊和壓縮偽影）時的魯棒性影響，是一個重要關注點。本文提供了一項全面的實證研究，評估了YOLO模型（從nano到extra-large規模）在多種精度格式下的魯棒性：FP32、FP16（TensorRT）、動態UINT8（ONNX）和靜態INT8（TensorRT）。我們提出並評估了一種針對靜態INT8 PTQ的退化感知校準策略，其中TensorRT校準過程會接觸到混合了乾淨和合成退化圖像的數據集。模型在COCO數據集上進行了基準測試，涵蓋七種不同的退化條件（包括各種類型和程度的噪聲、模糊、低對比度和JPEG壓縮）以及混合退化場景。結果表明，雖然靜態INT8 TensorRT引擎在乾淨數據上提供了顯著的加速（約1.5-3.3倍）且精度下降適中（約3-7% mAP50-95），但所提出的退化感知校準在大多數模型和退化條件下並未帶來一致且廣泛的魯棒性提升，相比於標準的乾淨數據校準。一個顯著的例外是在特定噪聲條件下，較大規模模型表現出改善，這表明模型容量可能影響此校準方法的有效性。這些發現凸顯了增強PTQ魯棒性的挑戰，並為在非受控環境中部署量化檢測器提供了見解。所有代碼和評估表格均可於https://github.com/AllanK24/QRID獲取。

English

Post-training quantization (PTQ) is crucial for deploying efficient object detection models, like YOLO, on resource-constrained devices. However, the impact of reduced precision on model robustness to real-world input degradations such as noise, blur, and compression artifacts is a significant concern. This paper presents a comprehensive empirical study evaluating the robustness of YOLO models (nano to extra-large scales) across multiple precision formats: FP32, FP16 (TensorRT), Dynamic UINT8 (ONNX), and Static INT8 (TensorRT). We introduce and evaluate a degradation-aware calibration strategy for Static INT8 PTQ, where the TensorRT calibration process is exposed to a mix of clean and synthetically degraded images. Models were benchmarked on the COCO dataset under seven distinct degradation conditions (including various types and levels of noise, blur, low contrast, and JPEG compression) and a mixed-degradation scenario. Results indicate that while Static INT8 TensorRT engines offer substantial speedups (~1.5-3.3x) with a moderate accuracy drop (~3-7% mAP50-95) on clean data, the proposed degradation-aware calibration did not yield consistent, broad improvements in robustness over standard clean-data calibration across most models and degradations. A notable exception was observed for larger model scales under specific noise conditions, suggesting model capacity may influence the efficacy of this calibration approach. These findings highlight the challenges in enhancing PTQ robustness and provide insights for deploying quantized detectors in uncontrolled environments. All code and evaluation tables are available at https://github.com/AllanK24/QRID.

物件偵測中輸入退化對量化之魯棒性

Quantization Robustness to Input Degradations for Object Detection

摘要

Support