Rethinking Image Aesthetics Assessment: Models, Datasets and Benchmarks
Authors: Shuai He, Yongchang Zhang, Rui Xie, Dongxiang Jiang, Anlong Ming
IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | we develop a large-scale benchmark (the most comprehensive thus far) by comparing 17 methods with TANet on three representative datasets: AVA, FLICKRAES and the proposed TAD66K, TANet achieves state-of-the-art performance on all three datasets. |
| Researcher Affiliation | Academia | Beijing University of Posts and Telecommunications {hs19951021, zhangyongchang, mibxr, jiangdx, mal}@bupt.edu.cn |
| Pseudocode | No | The paper illustrates the model architecture and equations but does not include any sections or figures labeled 'Pseudocode' or 'Algorithm'. |
| Open Source Code | Yes | Our work offers the community an opportunity to explore more challenging directions; the code, dataset and supplementary material are available at https://github.com/woshidandan/TANet. |
| Open Datasets | Yes | Specifically, 1) we elaborately build a novel dataset, called TAD66K, that contains 66K images covering 47 popular themes...; and two existing datasets, namely, Aesthetic Visual Analysis (AVA) [Murray et al., 2012] and Flickr Images with Aesthetics Annotation Dataset (FLICKR-AES) [Ren et al., 2017] |
| Dataset Splits | No | The paper discusses the creation of the TAD66K dataset and mentions using AVA and FLICKR-AES, and later refers to a 'test set', but it does not provide specific percentages, sample counts, or a detailed methodology for creating training, validation, and test splits required for reproduction. |
| Hardware Specification | Yes | Ablation studies of TANet on TAD66K (single 2080Ti). |
| Software Dependencies | No | The paper mentions general frameworks and languages used by various models in Table 2 (e.g., Py Torch, Tensorflow, MATLAB), but does not specify exact version numbers for these or other critical software dependencies required to reproduce their experiments. |
| Experiment Setup | No | The paper describes the model components and overall training approach, but it does not provide specific hyperparameter values (e.g., learning rate, batch size, number of epochs, optimizer settings) or detailed system-level training configurations in the main text. |