{"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/deep-attention-based-classification-network","title":"Deep attention-based classification network for robust depth prediction","arxiv_id":"1807.03959","date":"2018-07-11","proceeding":null,"authors":["Ruibo Li","Ke Xian","Chunhua Shen","Zhiguo Cao","Hao Lu","Lingxiao Hang"],"abstract":"In this paper, we present our deep attention-based classification (DABC)\nnetwork for robust single image depth prediction, in the context of the Robust\nVision Challenge 2018 (ROB 2018). Unlike conventional depth prediction, our\ngoal is to design a model that can perform well in both indoor and outdoor\nscenes with a single parameter set. However, robust depth prediction suffers\nfrom two challenging problems: a) How to extract more discriminative features\nfor different scenes (compared to a single scene)? b) How to handle the large\ndifferences of depth ranges between indoor and outdoor datasets? To address\nthese two problems, we first formulate depth prediction as a multi-class\nclassification task and apply a softmax classifier to classify the depth label\nof each pixel. We then introduce a global pooling layer and a channel-wise\nattention mechanism to adaptively select the discriminative channels of\nfeatures and to update the original features by assigning important channels\nwith higher weights. Further, to reduce the influence of quantization errors,\nwe employ a soft-weighted sum inference strategy for the final prediction.\nExperimental results on both indoor and outdoor datasets demonstrate the\neffectiveness of our method. It is worth mentioning that we won the 2-nd place\nin single image depth prediction entry of ROB 2018, in conjunction with IEEE\nConference on Computer Vision and Pattern Recognition (CVPR) 2018.","url_abs":"http://arxiv.org/abs/1807.03959v1","url_pdf":"http://arxiv.org/pdf/1807.03959v1.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":"deep-attention-based-classification-network","repo_url":"https://github.com/keerthi165/depthEstimation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"deep-attention","task_name":"Deep Attention"},{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"depth-prediction","task_name":"Depth Prediction"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"multi-class-classification","task_name":"Multi-class Classification"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"quantization","task_name":"Quantization"}],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.03959","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}