PAC-Net: Pairwise Aesthetic Comparison Network for Image Aesthetic Assessment

Ko, Keunsoo; Lee, Jun-Tae

doi:10.1109/icip.2018.8451621

Cited by 18 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The corresponding training and testing splits are also shown. [42] 0.708 ----RGNet [25] 0.710 ----PAC-Net [18] 0.837 ----Lee et al [21] 0.879 -0.1141 --HLA-GCN 0.899 0.9037 0.0980 0.0842 0.1093 stochastic gradient descent (SGD) following [42]. During training, we employ the layout-preserving warping introduced in Sec.…”

Section: Methodsmentioning

confidence: 99%

“…Each image is annotated by, on average, five people with integer scores ranging from 1 to 5. Following the previous work [18,19,21], we use the standard split with 8,500 images for training, 500 images from validation, and 1,000 images for testing. We normalize the score histogram to generate distributions and discard about 100 images that have only one rating following [47].…”

Section: Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

Hierarchical Layout-Aware Graph Convolutional Network for Unified Aesthetics Assessment

She

Lai

et al. 2021

2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Section: Datasetsmentioning

confidence: 99%

Hierarchical Layout-Aware Graph Convolutional Network for Unified Aesthetics Assessment

She

Lai

et al. 2021

2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

“…Each image is rated by about five people with integer scores ranging from 1 to 5. Following the previous work, 14 , 44 , 45 the standard split rule is used with 8500 images for training, 500 images from validation, and 1000 images for testing. The score histogram is normalized to generate distributions, and about 100 images that have only one rating is discarded following 39 …”

Section: Resultsmentioning

confidence: 99%

Another scale-guided parallel transformer for image aesthetic assessment

Shen

Zhang

et al. 2023

J. Electron. Imag.

View full text Add to dashboard Cite

Image aesthetic assessment (IAA) is a challenging task in computer vision fields, which aims to automatically evaluate image beauty by simulating human perception on image aesthetic. With the development of deep learning, although convolutional neural network (CNN)-based IAA approaches have achieved extraordinary progress, CNN experiences difficulty to capture long-distance relationships among visual elements. There is a strong correlation between image layout and image semantic information for image aesthetic. In order to solve this problem, an another scale-guided parallel transformer is proposed, including a multiscale local feature extractor (ME), a feature projection (FP), and an another scale-guided parallel feature fusion transformer (AST). The ME captures primary local features with classic ResNet at multiple scales. The FP performs dimension transformation on feature maps for each scale, which can obtain feature token and aesthetic token. The AST with two parallel transformer encoders is exploited to highlight the significant regions in the holistic image, in which the feature tokens and the aesthetic token from another scale are grouped together to obtain interscale guidance. The final score distribution is achieved by weighting multiple aesthetic tokens with learnable parameters for unified aesthetics assessment. Extensive experiments on two public datasets, including aesthetic visual analysis and aesthetics and attributes database, demonstrate that the proposed method outperforms the state-of-the-art methods across three different tasks.

show abstract

“…Existing works such as [17,62,94] leverage spatial pyramid pooling (SPP) [44] to avoid altering the resolution and aspect ratio of an input image. The SPP layer is able to pool arbitrarily-sized features 𝑥 and generate fixed-length outputs.…”

Section: Cnn-based Cpamentioning

confidence: 99%

AI-Empowered Persuasive Video Generation: A Survey

Liu¹,

Han²

2021

Preprint

View full text Add to dashboard Cite

Promotional videos are rapidly becoming a popular medium for persuading people to change their behaviours in many settings (e.g., online shopping, social enterprise initiatives). Today, such videos are often produced by professionals, which is a time-, labour-and cost-intensive undertaking. In order to produce such contents to support a large applications (e.g., e-commerce), the field of artificial intelligence (AI)-empowered persuasive video generation (AIPVG) has gained traction in recent years. This field is interdisciplinary in nature, which makes it challenging for new researchers to grasp. Currently, there is no comprehensive survey of AIPVG available. In this paper, we bridge this gap by reviewing key AI techniques that can be utilized to automatically generate persuasive videos. We offer a first-of-its-kind taxonomy which divides AIPVG into three major steps: 1) visual material understanding, which extracts information from the visual materials (VMs) relevant to the target of promotion; 2) visual storyline generation, which shortlists and arranges high-quality VMs into a sequence in order to compose a storyline with persuasive power; and 3) post-production, which involves background music generation and still image animation to enhance viewing experience. We also introduce the evaluation metrics and datasets commonly adopted in the field of AIPVG. We analyze the advantages and disadvantages of the existing works belonging to the above-mentioned steps, and discuss interesting potential future research directions.

show abstract

PAC-Net: Pairwise Aesthetic Comparison Network for Image Aesthetic Assessment

Cited by 18 publications

References 14 publications

Hierarchical Layout-Aware Graph Convolutional Network for Unified Aesthetics Assessment

Hierarchical Layout-Aware Graph Convolutional Network for Unified Aesthetics Assessment

Another scale-guided parallel transformer for image aesthetic assessment

AI-Empowered Persuasive Video Generation: A Survey

Contact Info

Product

Resources

About