{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/a-lamp-adaptive-layout-aware-multi-patch-deep","title":"A-Lamp: Adaptive Layout-Aware Multi-Patch Deep Convolutional Neural Network for Photo Aesthetic Assessment","arxiv_id":"1704.00248","date":"2017-04-02","proceeding":"CVPR 2017 7","authors":["Shuang Ma","Jing Liu","Chang Wen Chen"],"abstract":"Deep convolutional neural networks (CNN) have recently been shown to generate\npromising results for aesthetics assessment. However, the performance of these\ndeep CNN methods is often compromised by the constraint that the neural network\nonly takes the fixed-size input. To accommodate this requirement, input images\nneed to be transformed via cropping, warping, or padding, which often alter\nimage composition, reduce image resolution, or cause image distortion. Thus the\naesthetics of the original images is impaired because of potential loss of fine\ngrained details and holistic image layout. However, such fine grained details\nand holistic image layout is critical for evaluating an image's aesthetics. In\nthis paper, we present an Adaptive Layout-Aware Multi-Patch Convolutional\nNeural Network (A-Lamp CNN) architecture for photo aesthetic assessment. This\nnovel scheme is able to accept arbitrary sized images, and learn from both\nfined grained details and holistic image layout simultaneously. To enable\ntraining on these hybrid inputs, we extend the method by developing a dedicated\ndouble-subnet neural network structure, i.e. a Multi-Patch subnet and a\nLayout-Aware subnet. We further construct an aggregation layer to effectively\ncombine the hybrid features from these two subnets. Extensive experiments on\nthe large-scale aesthetics assessment benchmark (AVA) demonstrate significant\nperformance improvement over the state-of-the-art in photo aesthetic\nassessment.","url_abs":"http://arxiv.org/abs/1704.00248v1","url_pdf":"http://arxiv.org/pdf/1704.00248v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"aesthetics-quality-assessment","task_name":"Aesthetics Quality Assessment"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/aesthetics-quality-assessment-on-ava","task":"Aesthetics Quality Assessment","dataset":"AVA","model":"A-Lamp","rank_in_archive_order":2,"of":9,"metrics":{"Accuracy":"82.5%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.00248","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}