NeuFlow v2：邊緣裝置上高效率的光流估計

NeuFlow v2: High-Efficiency Optical Flow Estimation on Edge Devices

August 19, 2024

作者: Zhiyong Zhang, Aniket Gupta, Huaizu Jiang, Hanumant Singh

cs.AI

摘要

對於各種實際應用來說，實時高精度光流估計至關重要。儘管最近基於學習的光流方法取得了高精度，但通常伴隨著顯著的計算成本。本文提出了一種高效的光流方法，平衡了高精度和降低計算需求。在 NeuFlow v1 的基礎上，我們引入了新組件，包括更輕量級的主幹和快速細化模塊。這兩個模塊有助於保持計算需求輕便，同時提供接近最先進精度。相較於其他最先進方法，我們的模型實現了10倍至70倍的加速，同時在合成和真實世界數據上保持可比性能。在 Jetson Orin Nano 上，我們的模型能夠以超過20 FPS 的速度運行在 512x384 解析度圖像上。完整的訓練和評估代碼可在 https://github.com/neufieldrobotics/NeuFlow_v2 找到。

English

Real-time high-accuracy optical flow estimation is crucial for various real-world applications. While recent learning-based optical flow methods have achieved high accuracy, they often come with significant computational costs. In this paper, we propose a highly efficient optical flow method that balances high accuracy with reduced computational demands. Building upon NeuFlow v1, we introduce new components including a much more light-weight backbone and a fast refinement module. Both these modules help in keeping the computational demands light while providing close to state of the art accuracy. Compares to other state of the art methods, our model achieves a 10x-70x speedup while maintaining comparable performance on both synthetic and real-world data. It is capable of running at over 20 FPS on 512x384 resolution images on a Jetson Orin Nano. The full training and evaluation code is available at https://github.com/neufieldrobotics/NeuFlow_v2.