{"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/shape-inpainting-using-3d-generative","title":"Shape Inpainting using 3D Generative Adversarial Network and Recurrent Convolutional Networks","arxiv_id":"1711.06375","date":"2017-11-17","proceeding":"ICCV 2017 10","authors":["Weiyue Wang","Qiangui Huang","Suya You","Chao Yang","Ulrich Neumann"],"abstract":"Recent advances in convolutional neural networks have shown promising results\nin 3D shape completion. But due to GPU memory limitations, these methods can\nonly produce low-resolution outputs. To inpaint 3D models with semantic\nplausibility and contextual details, we introduce a hybrid framework that\ncombines a 3D Encoder-Decoder Generative Adversarial Network (3D-ED-GAN) and a\nLong-term Recurrent Convolutional Network (LRCN). The 3D-ED-GAN is a 3D\nconvolutional neural network trained with a generative adversarial paradigm to\nfill missing 3D data in low-resolution. LRCN adopts a recurrent neural network\narchitecture to minimize GPU memory usage and incorporates an Encoder-Decoder\npair into a Long Short-term Memory Network. By handling the 3D model as a\nsequence of 2D slices, LRCN transforms a coarse 3D shape into a more complete\nand higher resolution volume. While 3D-ED-GAN captures global contextual\nstructure of the 3D shape, LRCN localizes the fine-grained details.\nExperimental results on both real-world and synthetic data show reconstructions\nfrom corrupted models result in complete and high-resolution 3D objects.","url_abs":"http://arxiv.org/abs/1711.06375v1","url_pdf":"http://arxiv.org/pdf/1711.06375v1.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":[{"paper_slug":"shape-inpainting-using-3d-generative","repo_url":"https://github.com/fdevmsy/3d_shape_inpainting","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":null,"task_name":"GPU"},{"task_slug":null,"task_name":"Generative Adversarial Network"}],"methods":[{"method_slug":"memory-network","method_name":"Memory Network"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.06375","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}