arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.06504eess.IVcs.CV
收藏

一种用于计算高效的移动逆色调映射的混合量化网络

A Mixed Quantization Network for Computationally Efficient Mobile Inverse Tone Mapping

Samsung Research UK(三星英国研究院);Universitat Autònoma de Barcelona(巴塞罗那自治大学)

Juan Borrego-Carazo, Mete Ozay, Frederik Laboyrie, Paul Wisbey

摘要:

从单个低动态范围(LDR)图像恢复高动态范围(HDR)图像,即逆色调映射(ITM),由于在曝光过度和曝光不足的区域中缺乏信息而具有挑战性。当前的方法专门集中于训练高性能但计算效率低的ITM模型,这又阻碍了ITM模型在具有有限计算能力的资源受限环境(诸如边缘和移动终端应用)中的部署。为此,我们建议将深度神经网络的有效操作与一种新的混合量化方案相结合,以构建一种性能良好但计算效率高的混合量化网络(MQN),该网络可以在移动平台上执行单图像ITM。在消融研究中,我们探讨了使用不同的注意机制,量化方案和损失函数对ITM任务中MQN性能的影响。在比较分析中,使用MQN训练的ITM模型在基准数据集上的表现与最先进的方法不相上下。MQN模型在延迟方面提供了10倍的改进,在内存消耗方面提供了25倍的改进。

英文摘要:

Recovering a high dynamic range (HDR) image from a single low dynamic range (LDR) image, namely inverse tone mapping (ITM), is challenging due to the lack of information in over- and under-exposed regions. Current methods focus exclusively on training high-performing but computationally inefficient ITM models, which in turn hinder deployment of the ITM models in resource-constrained environments with limited computing power such as edge and mobile device applications. To this end, we propose combining efficient operations of deep neural networks with a novel mixed quantization scheme to construct a well-performing but computationally efficient mixed quantization network (MQN) which can perform single image ITM on mobile platforms. In the ablation studies, we explore the effect of using different attention mechanisms, quantization schemes, and loss functions on the performance of MQN in ITM tasks. In the comparative analyses, ITM models trained using MQN perform on par with the state-of-the-art methods on benchmark datasets. MQN models provide up to 10 times improvement on latency and 25 times improvement on memory consumption.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑