{"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/detnet-a-backbone-network-for-object","title":"DetNet: A Backbone network for Object Detection","arxiv_id":"1804.06215","date":"2018-04-17","proceeding":null,"authors":["Zeming Li","Chao Peng","Gang Yu","Xiangyu Zhang","Yangdong Deng","Jian Sun"],"abstract":"Recent CNN based object detectors, no matter one-stage methods like YOLO,\nSSD, and RetinaNe or two-stage detectors like Faster R-CNN, R-FCN and FPN are\nusually trying to directly finetune from ImageNet pre-trained models designed\nfor image classification. There has been little work discussing on the backbone\nfeature extractor specifically designed for the object detection. More\nimportantly, there are several differences between the tasks of image\nclassification and object detection. 1. Recent object detectors like FPN and\nRetinaNet usually involve extra stages against the task of image classification\nto handle the objects with various scales. 2. Object detection not only needs\nto recognize the category of the object instances but also spatially locate the\nposition. Large downsampling factor brings large valid receptive field, which\nis good for image classification but compromises the object location ability.\nDue to the gap between the image classification and object detection, we\npropose DetNet in this paper, which is a novel backbone network specifically\ndesigned for object detection. Moreover, DetNet includes the extra stages\nagainst traditional backbone network for image classification, while maintains\nhigh spatial resolution in deeper layers. Without any bells and whistles,\nstate-of-the-art results have been obtained for both object detection and\ninstance segmentation on the MSCOCO benchmark based on our DetNet~(4.8G FLOPs)\nbackbone. The code will be released for the reproduction.","url_abs":"http://arxiv.org/abs/1804.06215v2","url_pdf":"http://arxiv.org/pdf/1804.06215v2.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":"detnet-a-backbone-network-for-object","repo_url":"https://github.com/becauseofAI/DetNet-Keras","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"detnet-a-backbone-network-for-object","repo_url":"https://github.com/tsing-cv/DetNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"object-detection-1","task_name":"object-detection"},{"task_slug":null,"task_name":"valid"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"detnet","method_name":"DetNet"},{"method_slug":"dilated-bottleneck-block","method_name":"Dilated Bottleneck Block"},{"method_slug":"dilated-bottleneck-with-projection-block","method_name":"Dilated Bottleneck with Projection Block"},{"method_slug":"dilated-convolution","method_name":"Dilated Convolution"},{"method_slug":"fpn","method_name":"FPN"},{"method_slug":"faster-r-cnn","method_name":"Faster R-CNN"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"non-maximum-suppression","method_name":"Non Maximum Suppression"},{"method_slug":"rpn","method_name":"RPN"},{"method_slug":"randomhorizontalflip","method_name":"Random Horizontal Flip"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"roi-align","method_name":"RoIAlign"},{"method_slug":"roipool","method_name":"RoIPool"},{"method_slug":"sgd-with-momentum","method_name":"SGD with Momentum"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"step-decay","method_name":"Step Decay"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[{"slug":"detnet","name":"DetNet","full_name":"DetNet"}],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.06215","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}