{"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/automatic-adaptation-of-object-detectors-to","title":"Automatic adaptation of object detectors to new domains using self-training","arxiv_id":"1904.07305","date":"2019-04-15","proceeding":"CVPR 2019 6","authors":["Aruni RoyChowdhury","Prithvijit Chakrabarty","Ashish Singh","SouYoung Jin","Huaizu Jiang","Liangliang Cao","Erik Learned-Miller"],"abstract":"This work addresses the unsupervised adaptation of an existing object\ndetector to a new target domain. We assume that a large number of unlabeled\nvideos from this domain are readily available. We automatically obtain labels\non the target data by using high-confidence detections from the existing\ndetector, augmented with hard (misclassified) examples acquired by exploiting\ntemporal cues using a tracker. These automatically-obtained labels are then\nused for re-training the original model. A modified knowledge distillation loss\nis proposed, and we investigate several ways of assigning soft-labels to the\ntraining examples from the target domain. Our approach is empirically evaluated\non challenging face and pedestrian detection tasks: a face detector trained on\nWIDER-Face, which consists of high-quality images crawled from the web, is\nadapted to a large-scale surveillance data set; a pedestrian detector trained\non clear, daytime images from the BDD-100K driving data set is adapted to all\nother scenarios such as rainy, foggy, night-time. Our results demonstrate the\nusefulness of incorporating hard examples obtained from tracking, the advantage\nof using soft-labels via distillation loss versus hard-labels, and show\npromising performance as a simple method for unsupervised domain adaptation of\nobject detectors, with minimal dependence on hyper-parameters.","url_abs":"http://arxiv.org/abs/1904.07305v1","url_pdf":"http://arxiv.org/pdf/1904.07305v1.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":"automatic-adaptation-of-object-detectors-to","repo_url":"https://github.com/Gabriel-Macias/robust_frcnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"pedestrian-detection","task_name":"Pedestrian Detection"},{"task_slug":"unsupervised-domain-adaptation","task_name":"Unsupervised Domain Adaptation"}],"methods":[{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.07305","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1904.07305"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Gabriel-Macias/robust_frcnn","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":18},"by_repo_kind":{"listed":{"samples":18,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"b6396e65cc2d636e","entry":"class_loss_regr","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/losses.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/losses.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b6396e65cc2d636e"}},{"code_sha256_prefix":"372c139f6aad1ede","entry":"correct_pad","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/mobilenetv2.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/mobilenetv2.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"372c139f6aad1ede"}},{"code_sha256_prefix":"081f0e59aebb52e3","entry":"dense_block","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/densenet.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/densenet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"081f0e59aebb52e3"}},{"code_sha256_prefix":"b5afb710b31b739c","entry":"get_img_output_length","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/densenet.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/densenet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b5afb710b31b739c"}},{"code_sha256_prefix":"d3bdc6b05dbd86b1","entry":"identity_block","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/resnet.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/resnet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d3bdc6b05dbd86b1"}},{"code_sha256_prefix":"aaf50b014616278e","entry":"identity_block_td","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/resnet.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/resnet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"aaf50b014616278e"}},{"code_sha256_prefix":"c9b1379de8ad48db","entry":"js_distance","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"train_phase3_frcnn.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/train_phase3_frcnn.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c9b1379de8ad48db"}},{"code_sha256_prefix":"3f4bd2f243aba23a","entry":"kl_div","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"train_phase3_frcnn.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/train_phase3_frcnn.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3f4bd2f243aba23a"}},{"code_sha256_prefix":"fa214efc02584165","entry":"nn_base","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/mobilenetv1.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/mobilenetv1.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fa214efc02584165"}},{"code_sha256_prefix":"ffce07a1ea81a6dd","entry":"nn_base","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/mobilenetv1_05.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/mobilenetv1_05.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ffce07a1ea81a6dd"}},{"code_sha256_prefix":"bf44a7fd1c6968f3","entry":"nn_base","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/mobilenetv1_25.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/mobilenetv1_25.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"bf44a7fd1c6968f3"}},{"code_sha256_prefix":"7710b0425c1f28e8","entry":"nn_base","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/mobilenetv2.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/mobilenetv2.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7710b0425c1f28e8"}},{"code_sha256_prefix":"be60a1ca927bfdb5","entry":"rpn","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/mobilenetv1.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/mobilenetv1.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"be60a1ca927bfdb5"}},{"code_sha256_prefix":"dd3825b8d1875b76","entry":"rpn","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/mobilenetv1_05.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/mobilenetv1_05.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"dd3825b8d1875b76"}},{"code_sha256_prefix":"e8d8c25f96ea6956","entry":"rpn","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/mobilenetv1_25.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/mobilenetv1_25.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e8d8c25f96ea6956"}},{"code_sha256_prefix":"7fa7565671a5e2b3","entry":"rpn_loss_cls","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/losses.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/losses.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7fa7565671a5e2b3"}},{"code_sha256_prefix":"c0ceb156b9cb836c","entry":"rpn_loss_regr","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/losses.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/losses.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c0ceb156b9cb836c"}},{"code_sha256_prefix":"62b0b82803a67178","entry":"transition_block","repo":"Gabriel-Macias/robust_frcnn","repo_kind":"listed","path":"keras_frcnn/densenet.py","file_url":"https://github.com/Gabriel-Macias/robust_frcnn/blob/HEAD/keras_frcnn/densenet.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"62b0b82803a67178"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}