{"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/fine-tuning-image-conditional-diffusion","title":"Fine-Tuning Image-Conditional Diffusion Models is Easier than You Think","arxiv_id":"2409.11355","date":"2024-09-17","proceeding":null,"authors":["Gonzalo Martin Garcia","Karim Abou Zeid","Christian Schmidt","Daan de Geus","Alexander Hermans","Bastian Leibe"],"abstract":"Recent work showed that large diffusion models can be reused as highly precise monocular depth estimators by casting depth estimation as an image-conditional image generation task. While the proposed model achieved state-of-the-art results, high computational demands due to multi-step inference limited its use in many scenarios. In this paper, we show that the perceived inefficiency was caused by a flaw in the inference pipeline that has so far gone unnoticed. The fixed model performs comparably to the best previously reported configuration while being more than 200$\\times$ faster. To optimize for downstream task performance, we perform end-to-end fine-tuning on top of the single-step model with task-specific losses and get a deterministic model that outperforms all other diffusion-based depth and normal estimation models on common zero-shot benchmarks. We surprisingly find that this fine-tuning protocol also works directly on Stable Diffusion and achieves comparable performance to current state-of-the-art diffusion-based depth and normal estimation models, calling into question some of the conclusions drawn from prior works.","url_abs":"https://arxiv.org/abs/2409.11355v1","url_pdf":"https://arxiv.org/pdf/2409.11355v1.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":"fine-tuning-image-conditional-diffusion","repo_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"conditional-image-generation","task_name":"Conditional Image Generation"},{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"monocular-depth-estimation","task_name":"Monocular Depth Estimation"},{"task_slug":"surface-normal-estimation","task_name":"Surface Normal Estimation"},{"task_slug":"surface-normals-estimation","task_name":"Surface Normals Estimation"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/monocular-depth-estimation-on-nyu-depth-v2","task":"Monocular Depth Estimation","dataset":"NYU-Depth V2","model":"Marigold + E2E FT(zero-shot)","rank_in_archive_order":9,"of":85,"metrics":{"Delta < 1.25":"0.966","absolute relative error":"0.052"},"uses_additional_data":false},{"leaderboard":"/sota/surface-normals-estimation-on-ibims-1","task":"Surface Normals Estimation","dataset":"IBims-1","model":"Marigold + E2E FT(zero-shot)","rank_in_archive_order":1,"of":2,"metrics":{"% < 11.25":"69.9","Mean":"15.8"},"uses_additional_data":false},{"leaderboard":"/sota/surface-normals-estimation-on-nyu-depth-v2-1","task":"Surface Normals Estimation","dataset":"NYU Depth v2","model":"Marigold + E2E FT(zero-shot)","rank_in_archive_order":5,"of":6,"metrics":{"% < 11.25":"61.4","Mean Angle Error":"16.2"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2409.11355","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2409.11355"}},"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/VisualComputingInstitute/diffusion-e2e-ft","reach":{"status":"ok"}}],"summary":{"ran":16,"unverified":1},"by_repo_kind":{"official":{"samples":17,"ran":16,"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":17,"samples":[{"code_sha256_prefix":"3b234f415dac393b","entry":"compute_scale_and_shift_masked","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"training/util/loss.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/training/util/loss.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":"3b234f415dac393b"}},{"code_sha256_prefix":"b48c284df8bc63a7","entry":"define_loss","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/projects/baseline_normal/losses.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/projects/baseline_normal/losses.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":"b48c284df8bc63a7"}},{"code_sha256_prefix":"2c816eedd2f0474c","entry":"get_r_pitch","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/rotation.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/rotation.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":"2c816eedd2f0474c"}},{"code_sha256_prefix":"85d2c7386475ae37","entry":"get_r_roll","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/rotation.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/rotation.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":"85d2c7386475ae37"}},{"code_sha256_prefix":"14244d03bf3e5f48","entry":"get_r_yaw","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/rotation.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/rotation.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":"14244d03bf3e5f48"}},{"code_sha256_prefix":"07d8c13c1190a20a","entry":"intrins_crop","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/projection.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/projection.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":"07d8c13c1190a20a"}},{"code_sha256_prefix":"1218c2650fe6ffed","entry":"intrins_to_zero","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/projection.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/projection.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":"1218c2650fe6ffed"}},{"code_sha256_prefix":"f56119d671b23caa","entry":"intrins_zero_to","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/projection.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/projection.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":"f56119d671b23caa"}},{"code_sha256_prefix":"30d6db04f2c86d1d","entry":"l1_loss","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/projects/baseline_normal/losses.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/projects/baseline_normal/losses.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":"30d6db04f2c86d1d"}},{"code_sha256_prefix":"542e39255f62d87e","entry":"l2_loss","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/projects/baseline_normal/losses.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/projects/baseline_normal/losses.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":"542e39255f62d87e"}},{"code_sha256_prefix":"769dc6d3c6aa40de","entry":"load_checkpoint","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/utils.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/utils.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":"769dc6d3c6aa40de"}},{"code_sha256_prefix":"11eeeb13d2957ebf","entry":"pyramid_noise_like","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"GeoWizard/geowizard/models/geowizard_pipeline.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/GeoWizard/geowizard/models/geowizard_pipeline.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":"11eeeb13d2957ebf"}},{"code_sha256_prefix":"dc4ca4a94df3d89e","entry":"setup_custom_logger","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/utils.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/utils.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":"dc4ca4a94df3d89e"}},{"code_sha256_prefix":"e4738f6dce471506","entry":"tensor_to_numpy","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/visualize.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/visualize.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":"e4738f6dce471506"}},{"code_sha256_prefix":"2324fc8ae05d9f6d","entry":"txt_to_list","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/utils.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/utils.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":"2324fc8ae05d9f6d"}},{"code_sha256_prefix":"61cf14e229883558","entry":"unnormalize","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/visualize.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/visualize.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":"61cf14e229883558"}},{"code_sha256_prefix":"774236d466c851a3","entry":"normal_to_rgb","repo":"VisualComputingInstitute/diffusion-e2e-ft","repo_kind":"official","path":"DSINE/utils/visualize.py","file_url":"https://github.com/VisualComputingInstitute/diffusion-e2e-ft/blob/HEAD/DSINE/utils/visualize.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":"774236d466c851a3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}