{"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/depth-adaptive-deep-neural-network-for","title":"Depth Adaptive Deep Neural Network for Semantic Segmentation","arxiv_id":"1708.01818","date":"2017-08-05","proceeding":null,"authors":["Byeongkeun Kang","Yeejin Lee","Truong Q. Nguyen"],"abstract":"In this work, we present the depth-adaptive deep neural network using a depth\nmap for semantic segmentation. Typical deep neural networks receive inputs at\nthe predetermined locations regardless of the distance from the camera. This\nfixed receptive field presents a challenge to generalize the features of\nobjects at various distances in neural networks. Specifically, the\npredetermined receptive fields are too small at a short distance, and vice\nversa. To overcome this challenge, we develop a neural network which is able to\nadapt the receptive field not only for each layer but also for each neuron at\nthe spatial location. To adjust the receptive field, we propose the\ndepth-adaptive multiscale (DaM) convolution layer consisting of the adaptive\nperception neuron and the in-layer multiscale neuron. The adaptive perception\nneuron is to adjust the receptive field at each spatial location using the\ncorresponding depth information. The in-layer multiscale neuron is to apply the\ndifferent size of the receptive field at each feature space to learn features\nat multiple scales. The proposed DaM convolution is applied to two fully\nconvolutional neural networks. We demonstrate the effectiveness of the proposed\nneural networks on the publicly available RGB-D dataset for semantic\nsegmentation and the novel hand segmentation dataset for hand-object\ninteraction. The experimental results show that the proposed method outperforms\nthe state-of-the-art methods without any additional layers or\npre/post-processing.","url_abs":"http://arxiv.org/abs/1708.01818v2","url_pdf":"http://arxiv.org/pdf/1708.01818v2.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":"depth-adaptive-deep-neural-network-for","repo_url":"https://github.com/byeongkeun-kang/HOI-dataset","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"hand-segmentation","task_name":"Hand Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}