图像裁剪资料汇总

数据集

densely annotated (multiple crops in each image are annotated)

• CPC (2018) [link] （10797 images， each image has 24 crops）

• GAICD (2019) [link] （1236 images，86 crops for each image）

sparsely annotated (only the best crop in each image is annotated)

• KU-PCP（2018） [Paper] [Download link].

• ICDB/MSR-ICD (2013) [link] 950 images

• FLMS/HCDB (2014) [Download Images] [Download Crops] （500 images）

• FCDB (2017) [link] （1743 images, 1395 images are used for training and 348 for testing）

MSR-ICD：950 images which are originally from an image aesthetics assessment database（“Content-based photo quality assessment）。image categories, including animal, architecture, human, landscape, night, plant and man-made objects。Each image is carefully cropped by three expert photographers.

CPC: The CPC dataset (Wei et al. 2018) is the first densely annotated image cropping dataset. It contains 10797 images, and each image has 24 crops with multiple scores annotated by Amazon Mechanical Turk workers. We use the average annotated score for each crop as ground truth.

GAICD: The GAICD dataset (Zeng et al. 2019) is also a densely annotated dataset, but each image has more anno- tated crops than the CPC dataset. It has 1236 images and about 86 crops for each image.

FCDB&FLMS: The FCDB (Chen et al. 2017a) and FLMS (Fang et al. 2014) datasets only have annotations for the best crop in each image, which are given by ten annotators. FCDB contains 358 images, and FLMS contains 500 images. These two datasets are only used as the test set in the image cropping task

---

评价指标

GAICD：(SRCC) and AccK/N

IoU

BDE

---

论文

2013年

• author：Jianzhou Yan，Stephen Lin，Sing Bing Kang，Xiaoou Tang

• 单位：1The Chinese University of Hong Kong；2Microsoft Research Asia 3Microsoft Research

• Learning the Change for Automatic Image CroppinG

2016 CVPR

[Image-cropping]-Learning the Change for Automatic Image Cropping

• author： Jiansheng Chen； Gaocheng Bai；Shaoheng Liang；Zhengqin Li

• 单位：Department of Electronic Engineering, Tsinghua University, Beijing, China

• Automatic Image Cropping : A Computational Complexity Study

---

2017 Yi-Ling Chen WACV

• author：Yi-Ling Chen，Tzu-Wei Huang，Kai-Han Chang，Yu-Chen Tsai，Hwann-Tzong Chen， Bing-Yu Chen

• 单位：University of California, Davis ；National Taiwan University ；National Tsing Hua University

• Quantitative Analysis of Automatic Image Cropping Algorithms: A Dataset and Comparative Study（Flickr Cropping Dataset）

• yiling-chen/flickr-cropping-dataset: A dataset that includes photos downloaded from Flickr and annotations that indicates a local window representing a good composition. (github.com)

---

2017 Yi-Ling Chen MM

• author：Yi-Ling Chen，Tzu-Wei Huang，Kai-Han Chang，Yu-Chen Tsai，Hwann-Tzong Chen， Bing-Yu Chen

• 单位：University of California, Davis ；National Taiwan University ；National Tsing Hua University

• Learning to Compose with Professional Photographs on the Web

• remorsecs/pytorch-view-finding-network (github.com)

---

2018 CVPR

• author: Debang Li 1,2 , Huikai Wu 1,2 , Junge Zhang1,2 , Kaiqi Huang1,2,3

• 单位：CRIPAC & NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China 2 University of Chinese Academy of Sciences, Beijing, China 3 CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing, China

• title：A2-RL: Aesthetics Aware Reinforcement Learning for Image Cropping

• pjhool/TF-A2RL-Test: A2RL-Test (github.com)

• sunyasheng/a2rl_pytorch (github.com)

• Yasheng-Sun/a2rl-pytorch (github.com)

---

2018 CVPR

• author：Yubin Deng，Chen Change Loy，Xiaoou Tang

• 单位：CUHK-Sensetime Joint Laboratory, The Chinese University of Hong Kong

• title：Aesthetic-Driven Image Enhancement by Adversarial Learning

• dannysdeng/EnhanceGAN (github.com)

---

2018 CVPR

• author：Zijun Wei, Jianming Zhang, Xiaohui Shen, Zhe Lin, Radomir Mech, Minh Hoai, and Dimitris Samaras

• 单位：

• title：Good view hunting: Learning photo composition from dense view pairs

• Good View Hunting: Learning Photo Composition from Dense View Pairs (stonybrook.edu)

---

2018 TPAMI

• author：Wenguan Wang, Jianbing Shen, Senior Member, IEEE, and Haibin Ling

• 单位：Beijing Laboratory of Intelligent In- formation Technology, School of Computer Science, Beijing Institute of Technology；Department of Computer and Information Sciences, Temple University, Philadelphia, PA, USA

• title：A Deep Network Solution for Attention and Aesthetics Aware Photo Cropping

• 【Image-cropping】A Deep Network Solution for Attention and Aesthetics Aware Photo Cropping

---

2019 TIP

• author：Debang Li, Huikai Wu, Junge Zhang, and Kaiqi Huang

• 单位：CRIPAC & NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China 2 University of Chinese Academy of Sciences, Beijing, China 3 CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing, China

• title：Fast a3rl: Aesthetics-aware adversarial reinforcement learn- ing for image cropping.

---

2019 arixv

• author：Peng Lu, Hao Zhang, Xujun Peng, and Xiaofu Jin

• 单位：Beijing University of Posts and Telecommunications

• title：An End-to-End Neural Network for Image Cropping by Learning Composition from Aesthetic Photos

• CVBase-Bupt (github.com)

---

2020 TPAMI

• author：Hui Zeng, Lida Li, Zisheng Cao, and Lei Zhang, Fellow, IEEE

• 单位：H. Zeng, L. Li and L. Zhang are with the Department of Computing, The cslzhang}@comp.polyu.edu.hk). Hong Kong Polytechnic University, Hong

• title：Grid Anchor based Image Cropping: A New Benchmark and An Efficient Model

• HuiZeng/Grid-Anchor-based-Image-Cropping-Pytorch: PyTorch implementation of “Grid anchor based image cropping” (github.com)

• 【Image-cropping】Grid Anchor based Image Cropping: A New Benchmark and An Efficient Model

• 代码调试：GAIC代码调试

---

2020 CVPR

• author：Debang Li, Junge Zhang, and Kaiqi Huang

• 单位：CRIPAC & NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China 2 University of Chinese Academy of Sciences, Beijing, China 3 CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing, China

• title：Learning to learn cropping models for different aspect ratio requirements

---

2020 CVPR

• author：Debang Li, Junge Zhang, Kaiqi Huang, and Ming-Hsuan Yang

• 单位：CRIPAC & NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China 2 University of Chinese Academy of Sciences, Beijing, China 3 CAS Center for Excellence in Brain Science and Intelligence Technology, Beijing, China

• title：Composing good shots by exploiting mutual relations

---

2020 AAAI

• author：Yi Tu, Li Niu,Weijie Zhao, Dawei Cheng, and Liqing Zhang

• 单位：

• title：Image cropping with composition and saliency aware aesthetic score map

---

2021 tog

Aesthetic-guided outward image cropping | ACM Transactions on Graphics

---

2021 CVPR

• author: Chaoyi Hong†；Shuaiyuan Du†； Ke Xian†； Hao Lu† ；Zhiguo Cao†,* Weicai Zhong‡

• 单位：Key Laboratory of Image Processing and Intelligent Control, Ministry of Education School of Artificial Intelligence and Automation, Huazhong University of Science and Technology ‡Huawei CBG Consumer Cloud Service Search Product & Big Data Platform Department

• Composing Photos Like a Photographer code paper

• 笔记：Composing Photos Like a Photographer

• CACNet代码调试

2021 ICCV-Dissecting Image Crops

小组

P. Lu

P. Lu, H. Zhang, X. Peng, and X. Jin, ‘‘An end-to-end neural network for image cropping by learning composition from aesthetic photos,’’ ArXiv, vol. abs/1907.01432, 2019.

P. Lu, H. Zhang, X. Peng, and X. Peng, ‘‘Aesthetic guided deep regression network for image cropping,’’ Signal Process., Image Commun., vol. 77, pp. 1–10, Sep. 2019.

P. Lu, J. Liu, X. Peng, and X. Wang, ‘‘Weakly supervised real-time image cropping based on aesthetic distributions,’’ in Proc. 28th ACM Int. Conf. Multimedia (MM). New York, NY, USA: Association for Computing Machinery, 2020, pp. 120–128.

P. Lu, H. Zhang, X. Peng, and X. Jin, ‘‘Learning the relation between interested objects and aesthetic region for image cropping,’’ IEEE Trans. Multimedia, early access, Oct. 9, 2020

Debang Li

A2-RL: Aesthetics Aware Reinforcement Learning for Image Cropping（2018CVPR）

Fast a3rl: Aesthetics-aware adversarial reinforcement learn- ing for image cropping.（TIP2019）

Learning to learn cropping models for different aspect ratio requirements

Composing good shots by exploiting mutual relations

华科

Composing Photos Like a Photographer