{"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/pulse-self-supervised-photo-upsampling-via","title":"PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Models","arxiv_id":"2003.03808","date":"2020-03-08","proceeding":"CVPR 2020 6","authors":["Sachit Menon","Alexandru Damian","Shijia Hu","Nikhil Ravi","Cynthia Rudin"],"abstract":"The primary aim of single-image super-resolution is to construct high-resolution (HR) images from corresponding low-resolution (LR) inputs. In previous approaches, which have generally been supervised, the training objective typically measures a pixel-wise average distance between the super-resolved (SR) and HR images. Optimizing such metrics often leads to blurring, especially in high variance (detailed) regions. We propose an alternative formulation of the super-resolution problem based on creating realistic SR images that downscale correctly. We present an algorithm addressing this problem, PULSE (Photo Upsampling via Latent Space Exploration), which generates high-resolution, realistic images at resolutions previously unseen in the literature. It accomplishes this in an entirely self-supervised fashion and is not confined to a specific degradation operator used during training, unlike previous methods (which require supervised training on databases of LR-HR image pairs). Instead of starting with the LR image and slowly adding detail, PULSE traverses the high-resolution natural image manifold, searching for images that downscale to the original LR image. This is formalized through the \"downscaling loss,\" which guides exploration through the latent space of a generative model. By leveraging properties of high-dimensional Gaussians, we restrict the search space to guarantee realistic outputs. PULSE thereby generates super-resolved images that both are realistic and downscale correctly. We show proof of concept of our approach in the domain of face super-resolution (i.e., face hallucination). We also present a discussion of the limitations and biases of the method as currently implemented with an accompanying model card with relevant metrics. Our method outperforms state-of-the-art methods in perceptual quality at higher resolutions and scale factors than previously possible.","url_abs":"https://arxiv.org/abs/2003.03808v3","url_pdf":"https://arxiv.org/pdf/2003.03808v3.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":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/adamian98/pulse","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/Aal-Izz-Well/pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/Martion-z/Paddle-PULSE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"paddle","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/Martion-z/pytorch-Pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/b5071/pulse-rev","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/deeplearningnapratica/pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/eltechno/pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/jattardo1/pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/kingsj0405/pulse-any-image","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/krantirk/Self-Supervised-photo","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/marcin-laskowski/Pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/mlaskowski17/Pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/tg-bomze/Face-Depixelizer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/ytsheng/sketch2cat-pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/Mind23-2/MindCode-174","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"pulse-self-supervised-photo-upsampling-via","repo_url":"https://github.com/kainoj/keep-pulse","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"face-hallucination","task_name":"Face Hallucination"},{"task_slug":"hallucination","task_name":"Hallucination"},{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[{"method_slug":"pulse","method_name":"PULSE"}],"datasets_introduced":[],"methods_introduced":[{"slug":"pulse","name":"PULSE","full_name":"PULSE"}],"results":[{"leaderboard":"/sota/image-super-resolution-on-ffhq-256-x-256-4x","task":"Image Super-Resolution","dataset":"FFHQ 256 x 256 - 4x upscaling","model":"PULSE","rank_in_archive_order":11,"of":11,"metrics":{"PSNR":"15.74","SSIM":"0.37"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2003.03808","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2003.03808"}},"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/Martion-z/pytorch-Pulse","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/marcin-laskowski/Pulse","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/deeplearningnapratica/pulse","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Aal-Izz-Well/pulse","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/krantirk/Self-Supervised-photo","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tg-bomze/Face-Depixelizer","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Martion-z/Paddle-PULSE","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Mind23-2/MindCode-174","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ytsheng/sketch2cat-pulse","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mlaskowski17/Pulse","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/eltechno/pulse","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jattardo1/pulse","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kainoj/keep-pulse","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/adamian98/pulse","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/b5071/pulse-rev","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kingsj0405/pulse-any-image","reach":{"status":"ok"}}],"summary":{"unverified":3},"by_repo_kind":{"listed":{"samples":3,"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":"811b01b369c4bd3f","entry":"is_url","repo":"marcin-laskowski/Pulse","repo_kind":"listed","path":"drive.py","file_url":"https://github.com/marcin-laskowski/Pulse/blob/HEAD/drive.py","link_basis":"harvester_set","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":"811b01b369c4bd3f"}},{"code_sha256_prefix":"7bf55de655ef8dfa","entry":"open_url","repo":"marcin-laskowski/Pulse","repo_kind":"listed","path":"drive.py","file_url":"https://github.com/marcin-laskowski/Pulse/blob/HEAD/drive.py","link_basis":"harvester_set","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":"7bf55de655ef8dfa"}},{"code_sha256_prefix":"83889cc3c0ea78f3","entry":"upscale2d","repo":"marcin-laskowski/Pulse","repo_kind":"listed","path":"stylegan.py","file_url":"https://github.com/marcin-laskowski/Pulse/blob/HEAD/stylegan.py","link_basis":"harvester_set","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":"83889cc3c0ea78f3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}