{"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/learning-dynamic-memory-networks-for-object","title":"Learning Dynamic Memory Networks for Object Tracking","arxiv_id":"1803.07268","date":"2018-03-20","proceeding":"ECCV 2018 9","authors":["Tianyu Yang","Antoni B. Chan"],"abstract":"Template-matching methods for visual tracking have gained popularity recently\ndue to their comparable performance and fast speed. However, they lack\neffective ways to adapt to changes in the target object's appearance, making\ntheir tracking accuracy still far from state-of-the-art. In this paper, we\npropose a dynamic memory network to adapt the template to the target's\nappearance variations during tracking. An LSTM is used as a memory controller,\nwhere the input is the search feature map and the outputs are the control\nsignals for the reading and writing process of the memory block. As the\nlocation of the target is at first unknown in the search feature map, an\nattention mechanism is applied to concentrate the LSTM input on the potential\ntarget. To prevent aggressive model adaptivity, we apply gated residual\ntemplate learning to control the amount of retrieved memory that is used to\ncombine with the initial template. Unlike tracking-by-detection methods where\nthe object's information is maintained by the weight parameters of neural\nnetworks, which requires expensive online fine-tuning to be adaptable, our\ntracker runs completely feed-forward and adapts to the target's appearance\nchanges by updating the external memory. Moreover, unlike other tracking\nmethods where the model capacity is fixed after offline training --- the\ncapacity of our tracker can be easily enlarged as the memory requirements of a\ntask increase, which is favorable for memorizing long-term object information.\nExtensive experiments on OTB and VOT demonstrates that our tracker MemTrack\nperforms favorably against state-of-the-art tracking methods while retaining\nreal-time speed of 50 fps.","url_abs":"http://arxiv.org/abs/1803.07268v2","url_pdf":"http://arxiv.org/pdf/1803.07268v2.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":"learning-dynamic-memory-networks-for-object","repo_url":"https://github.com/skyoung/MemTrack","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"template-matching","task_name":"Template Matching"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"}],"methods":[{"method_slug":"dynamic-memory-network","method_name":"Dynamic Memory Network"},{"method_slug":"gru","method_name":"GRU"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"memory-network","method_name":"Memory Network"},{"method_slug":"speed","method_name":"SPEED"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.07268","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1803.07268"}},"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/skyoung/MemTrack","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":7},"by_repo_kind":{"official":{"samples":7,"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":"8228e128cba6b636","entry":"attention_read","repo":"skyoung/MemTrack","repo_kind":"official","path":"memnet/addressing.py","file_url":"https://github.com/skyoung/MemTrack/blob/HEAD/memnet/addressing.py","link_basis":"first_harvest_node","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":"8228e128cba6b636"}},{"code_sha256_prefix":"de17dc04cfed45b3","entry":"batch_conv","repo":"skyoung/MemTrack","repo_kind":"official","path":"model.py","file_url":"https://github.com/skyoung/MemTrack/blob/HEAD/model.py","link_basis":"first_harvest_node","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":"de17dc04cfed45b3"}},{"code_sha256_prefix":"0d47f81104cd3259","entry":"conv2d","repo":"skyoung/MemTrack","repo_kind":"official","path":"feature.py","file_url":"https://github.com/skyoung/MemTrack/blob/HEAD/feature.py","link_basis":"first_harvest_node","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":"0d47f81104cd3259"}},{"code_sha256_prefix":"ce4d099486b3c188","entry":"conv2d_bn_relu","repo":"skyoung/MemTrack","repo_kind":"official","path":"feature.py","file_url":"https://github.com/skyoung/MemTrack/blob/HEAD/feature.py","link_basis":"first_harvest_node","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":"ce4d099486b3c188"}},{"code_sha256_prefix":"ee9b47dc96eaf158","entry":"cosine_similarity","repo":"skyoung/MemTrack","repo_kind":"official","path":"memnet/addressing.py","file_url":"https://github.com/skyoung/MemTrack/blob/HEAD/memnet/addressing.py","link_basis":"first_harvest_node","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":"ee9b47dc96eaf158"}},{"code_sha256_prefix":"677fb42376c40578","entry":"extract_feature","repo":"skyoung/MemTrack","repo_kind":"official","path":"feature.py","file_url":"https://github.com/skyoung/MemTrack/blob/HEAD/feature.py","link_basis":"first_harvest_node","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":"677fb42376c40578"}},{"code_sha256_prefix":"ec2c36f05b2a0e9a","entry":"partition_vid","repo":"skyoung/MemTrack","repo_kind":"official","path":"build_tfrecords/build_data_vid.py","file_url":"https://github.com/skyoung/MemTrack/blob/HEAD/build_tfrecords/build_data_vid.py","link_basis":"first_harvest_node","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":"ec2c36f05b2a0e9a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}