{"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/squeezesegv2-improved-model-structure-and","title":"SqueezeSegV2: Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloud","arxiv_id":"1809.08495","date":"2018-09-22","proceeding":null,"authors":["Bichen Wu","Xuanyu Zhou","Sicheng Zhao","Xiangyu Yue","Kurt Keutzer"],"abstract":"Earlier work demonstrates the promise of deep-learning-based approaches for\npoint cloud segmentation; however, these approaches need to be improved to be\npractically useful. To this end, we introduce a new model SqueezeSegV2 that is\nmore robust to dropout noise in LiDAR point clouds. With improved model\nstructure, training loss, batch normalization and additional input channel,\nSqueezeSegV2 achieves significant accuracy improvement when trained on real\ndata. Training models for point cloud segmentation requires large amounts of\nlabeled point-cloud data, which is expensive to obtain. To sidestep the cost of\ncollection and annotation, simulators such as GTA-V can be used to create\nunlimited amounts of labeled, synthetic data. However, due to domain shift,\nmodels trained on synthetic data often do not generalize well to the real\nworld. We address this problem with a domain-adaptation training pipeline\nconsisting of three major components: 1) learned intensity rendering, 2)\ngeodesic correlation alignment, and 3) progressive domain calibration. When\ntrained on real data, our new model exhibits segmentation accuracy improvements\nof 6.0-8.6% over the original SqueezeSeg. When training our new model on\nsynthetic data using the proposed domain adaptation pipeline, we nearly double\ntest accuracy on real-world data, from 29.0% to 57.4%. Our source code and\nsynthetic dataset will be open-sourced.","url_abs":"http://arxiv.org/abs/1809.08495v1","url_pdf":"http://arxiv.org/pdf/1809.08495v1.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":"squeezesegv2-improved-model-structure-and","repo_url":"https://github.com/xuanyuzhou98/SqueezeSegV2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"BSD-2-Clause"}}],"tasks":[{"task_slug":"3d-semantic-segmentation","task_name":"3D Semantic Segmentation"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"point-cloud-segmentation","task_name":"Point Cloud Segmentation"},{"task_slug":"robust-3d-semantic-segmentation","task_name":"Robust 3D Semantic Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"unsupervised-domain-adaptation","task_name":"Unsupervised Domain Adaptation"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"dropout","method_name":"Dropout"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-semantic-segmentation-on-semantickitti","task":"3D Semantic Segmentation","dataset":"SemanticKITTI","model":"SqueezeSegV2","rank_in_archive_order":35,"of":45,"metrics":{"test mIoU":"39.7%"},"uses_additional_data":false},{"leaderboard":"/sota/robust-3d-semantic-segmentation-on","task":"Robust 3D Semantic Segmentation","dataset":"SemanticKITTI-C","model":"SqueezeSegV2 (64x2048)","rank_in_archive_order":21,"of":22,"metrics":{"mean Corruption Error (mCE)":"152.45%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.08495","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1809.08495"}},"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/xuanyuzhou98/SqueezeSegV2","reach":{"status":"ok","spdx":"BSD-2-Clause"}}],"summary":{"unverified":4},"by_repo_kind":{"listed":{"samples":4,"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":"23934dec57d44f3e","entry":"base_model_config","repo":"xuanyuzhou98/SqueezeSegV2","repo_kind":"listed","path":"src/config/config.py","file_url":"https://github.com/xuanyuzhou98/SqueezeSegV2/blob/HEAD/src/config/config.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-2-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"23934dec57d44f3e"}},{"code_sha256_prefix":"39ffc1ab87f2c16b","entry":"bgr_to_rgb","repo":"xuanyuzhou98/SqueezeSegV2","repo_kind":"listed","path":"src/utils/util.py","file_url":"https://github.com/xuanyuzhou98/SqueezeSegV2/blob/HEAD/src/utils/util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-2-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"39ffc1ab87f2c16b"}},{"code_sha256_prefix":"d98674c6c141b363","entry":"conf_error_rate_at_thresh_fn","repo":"xuanyuzhou98/SqueezeSegV2","repo_kind":"listed","path":"src/utils/util.py","file_url":"https://github.com/xuanyuzhou98/SqueezeSegV2/blob/HEAD/src/utils/util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-2-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"d98674c6c141b363"}},{"code_sha256_prefix":"9af1c222357f80b5","entry":"visualize_seg","repo":"xuanyuzhou98/SqueezeSegV2","repo_kind":"listed","path":"src/utils/util.py","file_url":"https://github.com/xuanyuzhou98/SqueezeSegV2/blob/HEAD/src/utils/util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-2-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"9af1c222357f80b5"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}