{"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/arxiv-2512-03210","title":"Flux4D: Flow-based Unsupervised 4D Reconstruction","arxiv_id":"2512.03210","date":"2025-12-02","proceeding":"NeurIPS","authors":["Jingkang Wang","Henry Che","Yun Chen","Ze Yang","Lily Goli","Sivabalan Manivasagam","Raquel Urtasun"],"abstract":"Reconstructing large-scale dynamic scenes from visual observations is a fundamental challenge in computer vision, with critical implications for robotics and autonomous systems. While recent differentiable rendering methods such as Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have achieved impressive photorealistic reconstruction, they suffer from scalability limitations and require annotations to decouple actor motion. Existing self-supervised methods attempt to eliminate explicit annotations by leveraging motion cues and geometric priors, yet they remain constrained by per-scene optimization and sensitivity to hyperparameter tuning. In this paper, we introduce Flux4D, a simple and scalable framework for 4D reconstruction of large-scale dynamic scenes. Flux4D directly predicts 3D Gaussians and their motion dynamics to reconstruct sensor observations in a fully unsupervised manner. By adopting only photometric losses and enforcing an \"as static as possible\" regularization, Flux4D learns to decompose dynamic elements directly from raw data without requiring pre-trained supervised models or foundational priors simply by training across many scenes. Our approach enables efficient reconstruction of dynamic scenes within seconds, scales effectively to large datasets, and generalizes well to unseen environments, including rare and unknown objects. Experiments on outdoor driving datasets show Flux4D significantly outperforms existing methods in scalability, generalization, and reconstruction quality.","url_abs":"https://arxiv.org/abs/2512.03210","url_pdf":"https://arxiv.org/pdf/2512.03210","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2512.03210","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2512.03210"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/NVlabs/EmerNeRF","reach":{"status":"ok","spdx":"NOASSERTION"}},{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/ziyc/drivestudio","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"ran_honours":1,"unverified":20},"by_repo_kind":{"found_in_text":{"samples":22,"ran":2,"repositories":2}},"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":9,"samples":[{"code_sha256_prefix":"8d68829e9ebe344d","entry":"get_embedder","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/modules.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/modules.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8d68829e9ebe344d"}},{"code_sha256_prefix":"57acaa6d628ba9f8","entry":"random_quat_tensor","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/gaussians/basics.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/gaussians/basics.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"57acaa6d628ba9f8"}},{"code_sha256_prefix":"935ce040cd9cf66c","entry":"binary_cross_entropy","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/losses.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/losses.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":"935ce040cd9cf66c"}},{"code_sha256_prefix":"46ff1acb735d62b5","entry":"blockPrinting","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/human_body.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/human_body.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":"46ff1acb735d62b5"}},{"code_sha256_prefix":"f0ec575e1a589ea4","entry":"build_scheduler_from_cfg","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"builders.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/builders.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f0ec575e1a589ea4"}},{"code_sha256_prefix":"37e6686ebde165ef","entry":"compute_psnr","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"datasets/metrics.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/datasets/metrics.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"37e6686ebde165ef"}},{"code_sha256_prefix":"7525a1bc40362d83","entry":"compute_range_image_cartesian","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"datasets/waymo_preprocess.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/datasets/waymo_preprocess.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"7525a1bc40362d83"}},{"code_sha256_prefix":"293991ba7c0526ca","entry":"compute_ssim","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"datasets/metrics.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/datasets/metrics.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"293991ba7c0526ca"}},{"code_sha256_prefix":"5768489744759311","entry":"compute_valid_depth_rmse","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"datasets/metrics.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/datasets/metrics.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"5768489744759311"}},{"code_sha256_prefix":"68e224cec422991b","entry":"contract","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/gaussians/deformgs.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/gaussians/deformgs.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":"68e224cec422991b"}},{"code_sha256_prefix":"96fd24a72d00adf0","entry":"dirac_delta_approx","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"loss/base.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/loss/base.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"96fd24a72d00adf0"}},{"code_sha256_prefix":"4d01d3ba48fe3cb9","entry":"get_linear_noise_func","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/gaussians/deformgs.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/gaussians/deformgs.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":"4d01d3ba48fe3cb9"}},{"code_sha256_prefix":"f6b2097bb697bae6","entry":"get_numpy","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/video_utils.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/video_utils.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":"f6b2097bb697bae6"}},{"code_sha256_prefix":"48923d99fae91824","entry":"get_predefined_human_rest_pose","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/human_body.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/human_body.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":"48923d99fae91824"}},{"code_sha256_prefix":"68f7385b9146d55a","entry":"interpolate_quats","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/gaussians/basics.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/gaussians/basics.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":"68f7385b9146d55a"}},{"code_sha256_prefix":"caa9641addd26530","entry":"non_zero_mean","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/video_utils.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/video_utils.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":"caa9641addd26530"}},{"code_sha256_prefix":"10f6c504485a371d","entry":"normalize_depth","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"loss/base.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/loss/base.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"10f6c504485a371d"}},{"code_sha256_prefix":"7ef271bf33d77004","entry":"quat_mult","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/gaussians/basics.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/gaussians/basics.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":"7ef271bf33d77004"}},{"code_sha256_prefix":"d64472f8d41f2419","entry":"reduce","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/losses.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/losses.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":"d64472f8d41f2419"}},{"code_sha256_prefix":"1c1e0d270ab71346","entry":"safe_binary_cross_entropy","repo":"ziyc/drivestudio","repo_kind":"found_in_text","path":"models/losses.py","file_url":"https://github.com/ziyc/drivestudio/blob/HEAD/models/losses.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":"1c1e0d270ab71346"}},{"code_sha256_prefix":"8cda114e6c7d9f38","entry":"voxel_coords_to_world_coords","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"datasets/utils.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/datasets/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"8cda114e6c7d9f38"}},{"code_sha256_prefix":"bd96c5131255d2f3","entry":"world_coords_to_voxel_coords","repo":"NVlabs/EmerNeRF","repo_kind":"found_in_text","path":"datasets/utils.py","file_url":"https://github.com/NVlabs/EmerNeRF/blob/HEAD/datasets/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"bd96c5131255d2f3"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CV","source":"arxiv_api"},"syntology_extracted_results":null}