.png){kind=icon}
返回
Image resizing by reconstruction from deep features
.png){kind=icon}
OA
AI
DOI：10.1007/s41095-021-0216-x.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Traditional image resizing methods usually work in pixel space and use various saliency measures. The challenge is to adjust the image shape while trying to preserve important content. In this paper we perform image resizing in feature space using the deep layers of a neural network containing rich important semantic information. We directly adjust the image feature maps, extracted from a pre-trained classification network, and reconstruct the resized image using neural-network based optimization. This novel approach leverages the hierarchical encoding of the network, and in particular, the high-level discriminative power of its deeper layers, that can recognize semantic regions and objects, thereby allowing maintenance of their aspect ratios. Our use of reconstruction from deep features results in less noticeable artifacts than use of imagespace resizing operators. We evaluate our method on benchmarks, compare it to alternative approaches, and demonstrate its strengths on challenging images.
Keyword:
image retargeting
reconstruction
deep seam carving
image resizing
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
18.3
论文数:
315
被引数:
2.6K
机构
引用论文
ARVino - outdoor augmented reality visualisation of viticulture GIS dataARVino-葡萄栽培GIS数据的户外增强现实可视化OAAI
OAAI分享收藏Serum factor in Miller-Fisher variant of Guillain-Barré syndrome and neurotransmitter release.png)
The Lancet
The LancetIF0
PREAI
分享收藏Simultaneous determination of some common food dyes in commercial products by digital image analysisOAAI
OAAI分享收藏
查看更多内容