{"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/elysium-exploring-object-level-perception-in","title":"Elysium: Exploring Object-level Perception in Videos via MLLM","arxiv_id":"2403.16558","date":"2024-03-25","proceeding":null,"authors":["Han Wang","Yanjie Wang","YongJie Ye","Yuxiang Nie","Can Huang"],"abstract":"Multi-modal Large Language Models (MLLMs) have demonstrated their ability to perceive objects in still images, but their application in video-related tasks, such as object tracking, remains understudied. This lack of exploration is primarily due to two key challenges. Firstly, extensive pretraining on large-scale video datasets is required to equip MLLMs with the capability to perceive objects across multiple frames and understand inter-frame relationships. Secondly, processing a large number of frames within the context window of Large Language Models (LLMs) can impose a significant computational burden. To address the first challenge, we introduce ElysiumTrack-1M, a large-scale video dataset supported for three tasks: Single Object Tracking (SOT), Referring Single Object Tracking (RSOT), and Video Referring Expression Generation (Video-REG). ElysiumTrack-1M contains 1.27 million annotated video frames with corresponding object boxes and descriptions. Leveraging this dataset, we conduct training of MLLMs and propose a token-compression model T-Selector to tackle the second challenge. Our proposed approach, Elysium: Exploring Object-level Perception in Videos via MLLM, is an end-to-end trainable MLLM that attempts to conduct object-level tasks in videos without requiring any additional plug-in or expert models. All codes and datasets are available at https://github.com/Hon-Wong/Elysium.","url_abs":"https://arxiv.org/abs/2403.16558v2","url_pdf":"https://arxiv.org/pdf/2403.16558v2.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":"elysium-exploring-object-level-perception-in","repo_url":"https://github.com/hon-wong/elysium","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"referring-expression","task_name":"Referring Expression"},{"task_slug":"referring-expression-comprehension","task_name":"Referring Expression Comprehension"},{"task_slug":"referring-expression-generation","task_name":"Referring expression generation"},{"task_slug":"video-question-answering","task_name":"Video Question Answering"},{"task_slug":"visual-object-tracking","task_name":"Visual Object Tracking"},{"task_slug":"zero-shot-single-object-tracking","task_name":"Zero-Shot Single Object Tracking"},{"task_slug":"zeroshot-video-question-answer","task_name":"Zero-Shot Video Question Answer"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/zero-shot-single-object-tracking-on-lasot","task":"Zero-Shot Single Object Tracking","dataset":"LaSOT","model":"Elysium","rank_in_archive_order":2,"of":2,"metrics":{"AUC":"56.1","Normalized Precision":"61.0","Precision":"50.1"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-activitynet","task":"Zero-Shot Video Question Answer","dataset":"ActivityNet-QA","model":"Elysium","rank_in_archive_order":23,"of":28,"metrics":{"Accuracy":"43.4","Confidence Score":"2.9"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-msrvtt-qa","task":"Zero-Shot Video Question Answer","dataset":"MSRVTT-QA","model":"Elysium","rank_in_archive_order":3,"of":30,"metrics":{"Accuracy":"67.5","Confidence Score":"3.2"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-msvd-qa","task":"Zero-Shot Video Question Answer","dataset":"MSVD-QA","model":"Elysium","rank_in_archive_order":10,"of":28,"metrics":{"Accuracy":"75.8","Confidence Score":"3.7"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-tgif-qa","task":"Zero-Shot Video Question Answer","dataset":"TGIF-QA","model":"Elysium","rank_in_archive_order":11,"of":14,"metrics":{"Accuracy":"66.6","Confidence Score":"3.6"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.16558","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.16558"}},"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/hon-wong/elysium","reach":{"status":"ok"}}],"summary":{"ran":7,"unverified":1},"by_repo_kind":{"official":{"samples":8,"ran":7,"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":8,"samples":[{"code_sha256_prefix":"03ea4ef71d990d69","entry":"b2str","repo":"hon-wong/elysium","repo_kind":"official","path":"eval/merge_result.py","file_url":"https://github.com/hon-wong/elysium/blob/HEAD/eval/merge_result.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"03ea4ef71d990d69"}},{"code_sha256_prefix":"14a5b06381b76869","entry":"build_adapter","repo":"hon-wong/elysium","repo_kind":"official","path":"models/rantselector.py","file_url":"https://github.com/hon-wong/elysium/blob/HEAD/models/rantselector.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"14a5b06381b76869"}},{"code_sha256_prefix":"dc5f7a793bd5c8b8","entry":"build_projector","repo":"hon-wong/elysium","repo_kind":"official","path":"models/build_vit.py","file_url":"https://github.com/hon-wong/elysium/blob/HEAD/models/build_vit.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dc5f7a793bd5c8b8"}},{"code_sha256_prefix":"66f1707e5716c275","entry":"calc_err_center","repo":"hon-wong/elysium","repo_kind":"official","path":"eval/otb.py","file_url":"https://github.com/hon-wong/elysium/blob/HEAD/eval/otb.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"66f1707e5716c275"}},{"code_sha256_prefix":"3079b3e29bfecb3d","entry":"calc_iou_overlap","repo":"hon-wong/elysium","repo_kind":"official","path":"eval/otb.py","file_url":"https://github.com/hon-wong/elysium/blob/HEAD/eval/otb.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3079b3e29bfecb3d"}},{"code_sha256_prefix":"30aa0d18d04fc6fb","entry":"calc_seq_err_robust","repo":"hon-wong/elysium","repo_kind":"official","path":"eval/otb.py","file_url":"https://github.com/hon-wong/elysium/blob/HEAD/eval/otb.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"30aa0d18d04fc6fb"}},{"code_sha256_prefix":"7f7a40ee6ae5eb93","entry":"parse_box_from_raw_text","repo":"hon-wong/elysium","repo_kind":"official","path":"eval/merge_result.py","file_url":"https://github.com/hon-wong/elysium/blob/HEAD/eval/merge_result.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7f7a40ee6ae5eb93"}},{"code_sha256_prefix":"69245b70a5e8929e","entry":"create_clip_vit","repo":"hon-wong/elysium","repo_kind":"official","path":"models/build_vit.py","file_url":"https://github.com/hon-wong/elysium/blob/HEAD/models/build_vit.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"69245b70a5e8929e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}