{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/hierarchical-interaction-network-for-video","title":"Hierarchical interaction network for video object segmentation from referring expressions","arxiv_id":null,"date":"2021-11-22","proceeding":"British Machine Vision Conference 2021 11","authors":["Zhao Yang","Yansong Tang","Luca Bertinetto","Hengshuang Zhao","Philip Torr"],"abstract":"In this paper, we investigate the problem of video object segmentation from referring expressions (VOSRE). Conventional methods typically perform multi-modal fusion based on linguistic features and the visual features extracted from the top layer of the visual encoder, which limits these models' ability to represent multi-modal inputs at different semantic and spatial granularity levels. To address this issue, we present an end-to-end hierarchical interaction network (HINet) for the VOSRE problem. Our model leverages the feature pyramid produced by the visual encoder to generate multiple levels of multi-modal features. This allows more flexible representation of various linguistic concepts (e.g., object attributes and categories) in different levels of the multi-modal features. Moreover, we further extract signals of moving objects from optical flow input, and utilize them as complementary cues for highlighting the referent and suppressing the background with a motion gating mechanism. In contrast to previous methods, this strategy allows our model to make online predictions without requiring the whole video as input. Despite its simplicity, our proposed HINet improves over the previous state of the art on the DAVIS-16, DAVIS-17, and J-HMDB datasets for the VOSRE task, demonstrating its effectiveness and generality.","url_abs":"https://www.bmvc2021-virtualconference.com/conference/papers/paper_0386.html","url_pdf":"https://www.bmvc2021-virtualconference.com/assets/papers/0386.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":[],"tasks":[{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"},{"task_slug":"referring-expression-segmentation","task_name":"Referring Expression Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"video-object-segmentation","task_name":"Video Object Segmentation"},{"task_slug":"video-semantic-segmentation","task_name":"Video Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/referring-expression-segmentation-on-a2d","task":"Referring Expression Segmentation","dataset":"A2D Sentences","model":"HINet","rank_in_archive_order":23,"of":27,"metrics":{"IoU mean":"0.529","IoU overall":"0.679","Precision@0.5":"0.611","Precision@0.6":"0.559","Precision@0.7":"0.486","Precision@0.8":"0.342","Precision@0.9":"0.12"},"uses_additional_data":false},{"leaderboard":"/sota/referring-expression-segmentation-on-a2d","task":"Referring Expression Segmentation","dataset":"A2D Sentences","model":"RefVOS","rank_in_archive_order":24,"of":27,"metrics":{"IoU mean":"0.497","IoU overall":"0.672","Precision@0.5":"0.578","Precision@0.6":"0.534","Precision@0.7":"0.456","Precision@0.8":"0.311","Precision@0.9":"0.093"},"uses_additional_data":false},{"leaderboard":"/sota/referring-expression-segmentation-on-davis","task":"Referring Expression Segmentation","dataset":"DAVIS 2017 (val)","model":"HINet","rank_in_archive_order":10,"of":18,"metrics":{"J&F 1st frame":"50.2","J&F Full video":"47.9"},"uses_additional_data":false},{"leaderboard":"/sota/referring-expression-segmentation-on-j-hmdb","task":"Referring Expression Segmentation","dataset":"J-HMDB","model":"HINet","rank_in_archive_order":18,"of":21,"metrics":{"IoU mean":"0.627","IoU overall":"0.652","Precision@0.5":"0.819","Precision@0.6":"0.736","Precision@0.7":"0.542","Precision@0.8":"0.168","Precision@0.9":"0.4"},"uses_additional_data":false},{"leaderboard":"/sota/referring-expression-segmentation-on-j-hmdb","task":"Referring Expression Segmentation","dataset":"J-HMDB","model":"RefVOS","rank_in_archive_order":21,"of":21,"metrics":{"IoU mean":"0.568","IoU overall":"0.606","Precision@0.5":"0.731","Precision@0.6":"0.62","Precision@0.7":"0.392","Precision@0.8":"0.088","Precision@0.9":"0.0"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers","held":null,"reason":"no_arxiv_id","checked_against":null,"by_repo":[],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":"the archive holds no code link for this paper","arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"06d43325e3040a442d5deb35ff37ee87dcffa89f480dcc9414e6191364cb295f","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}