Hierarchical Fine-Grained Image Forgery Detection and Localization
Open accessThe DOI matches its Crossref record; 35 of the 35 references it cites were found in a registry or an index.
Abstract
Differences in forgery attributes of images generated in CNN-synthesized and image-editing domains are large, and such differences make a unified image forgery detection and localization (IFDL) challenging. To this end, we present a hierarchical fine-grained formulation for IFDL representation learning. Specifically, we first represent forgery attributes of a manipulated image with multiple labels at different levels. Then we perform fine-grained classification at these levels using the hierarchical dependency between them. As a result, the algorithm is encouraged to learn both comprehensive features and inherent hierarchical nature of different forgery attributes, thereby improving the IFDL representation. Our proposed IFDL framework contains three components: multi-branch feature extractor, localization and classification modules. Each branch of the feature extractor learns to classify forgery attributes at one level, while localization and classification modules segment the pixel-level forgery region and detect image-level forgery, respectively. Lastly, we construct a hierarchical fine-grained dataset to facilitate our study. We demonstrate the effectiveness of our method on 7 different benchmarks, for both tasks of IFDL and forgery attribute classification. Our source code and dataset can be found: github.com/CHELSEA234/HiFi-IFDL.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
High-Resolution Image Synthesis with Latent Diffusion Models
Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01042 · Cited by 20,302
A Style-Based Generator Architecture for Generative Adversarial Networks
Karras, T., Laine, S., & Aila, T. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00453 · Cited by 11,714
Wang, X., Girshick, R., Gupta, A., & He, K. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00813 · Cited by 11,596
Denoising Diffusion Probabilistic Models
Ho, J., Jain, A., & Abbeel, P. (2020) · arXiv · doi:10.48550/arxiv.2006.11239 · Cited by 8,712
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Karras, T., Aila, T., Laine, S., & Lehtinen, J. (2017) · arXiv · doi:10.48550/arxiv.1710.10196 · Cited by 4,727
Self-Attention Generative Adversarial Networks
Zhang, H., Goodfellow, I., Metaxas, D., & Odena, A. (2018) · arXiv · doi:10.48550/arxiv.1805.08318 · Cited by 3,353 · cited 1×
“To model the dependency and interactions of pixels on the large spatial area, the localization module employs the self-attention mechanism [65, 72].”
FaceForensics++: Learning to Detect Manipulated Facial Images
Rossler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J., & Niessner, M. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00009 · Cited by 3,314 · cited 4×
“The real image dataset is the combination of LSUN [75], CelebaHQ [37], FFHQ [33], AFHQ [12], MSCOCO [41] and real face images in face forensics [58].”
BackgroundMethodInfluentialPhotorealistic Text-to-Image Diffusion Models with Deep Language Understanding
Saharia, C., Chan, W., Saxena, S., Li, L., Whang, J., Denton, E., Ghasemipour, S. K. S., Ayan, B. K., Mahdavi, S. S., Lopes, R. G., et al. (2022) · arXiv · doi:10.48550/arxiv.2205.11487 · Cited by 2,528 · cited 2×
“Firstly, CNN-synthesized images made extraordinary leaps culminating in recent synthesis methods—Dall E [50] or Google ImageN [54]—based on diffusion models (DDPM) [22], which even generates realistic videos from text [21, 57].”
Hierarchical Text-Conditional Image Generation with CLIP Latents
Ramesh, A., Dhariwal, P., Nichol, A., Chu, C., & Chen, M. (2022) · arXiv · doi:10.48550/arxiv.2204.06125 · Cited by 2,511
Denoising Diffusion Implicit Models
Song, J., Meng, C., & Ermon, S. (2020) · arXiv · doi:10.48550/arxiv.2010.02502 · Cited by 2,507 · cited 6×
“c Tr ue L ab el Real Splice Inpaint [44] [44] [38] [12] [40] [31] [32] [25] [64] [52] [57]”
BackgroundMethodInfluential
Research works printed but not linked: 42
- ruff, 2018, Deep one-class classification, ICML
- 10.1109/FG57933.2023.10042500 · DOI · Paper page
- 10.1007/978-981-16-7621-5 · DOI
- 10.1109/ACCESS.2019.2899367 · DOI · Paper page
- ricker, 2022, Towards the detection of diffusion model deepfakes, ArXiv Preprint
- 10.1109/ICCV48922.2021.01392 · DOI · Paper page
- 10.1109/TPAMI.2022.3180556 · DOI · Paper page
- dufour, 2019, Deepfakes detection dataset by google & jigsaw
- 10.1109/ChinaSIP.2013.6625374 · DOI · Paper page
- 10.1145/882262.882269 · DOI · Paper page
- 10.1109/WACVW50321.2020.9096940 · DOI · Paper page
- ng, 2009, Columbia image splicing detection evaluation dataset, DVMM lab Columbia Univ CalPhotos Digit Libr
- 10.1109/ICASSP.2018.8462585 · DOI · Paper page
- 10.1109/TCSVT.2022.3189545 · DOI · Paper page
- 10.1007/978-3-031-19781-9_9 · DOI
- 10.1109/CVPR52729.2023.01188 · DOI
- 10.1016/B978-0-08-051581-6.50065-9 · DOI · Paper page
- 10.1001/jamanetworkopen.2020.22199 · DOI
- 2022, Infodemic-world Health Organization
- asnani, 2021, Reverse engineering of generative models: Inferring model hyperparameters from generated images, ArXiv Preprint
- 10.1109/CVPR52688.2022.01495 · DOI
- lin, 2014, Microsoft coco: Common objects in context, ECCV
- 10.1007/s11263-022-01617-5 · DOI · Paper page
- 10.1109/CVPR42600.2020.00559 · DOI · Paper page
- karras, 2021, Alias-free generative adversarial networks, NeurIPS
- 10.1609/aaai.v34i07.7007 · DOI · Paper page
- kingma, 2014, Auto-encoding variational bayes, ICLRE
- 10.1109/CVPR.2018.00116 · DOI · Paper page
- 2016, NIST Nist Nimble 2016 Datasets
- 0, Survey More americans get news from internet than newspapers or radio
- li, 2022, Image-to-image translation via hierarchical style disentanglement, CVPR
- 10.1109/ICCV.2019.00765 · DOI · Paper page
- yu, 2015, Lsun: Construction of a large-scale image dataset using deep learning with humans in the loop, ArXiv Preprint
- 10.1109/CVPR.2019.00977 · DOI · Paper page
- 10.1007/978-3-030-01231-1_11 · DOI · Paper page
- hu, 2020, Span: spatial pyramid attention network for image manipulation localization, ECCV
- hsu, 2021, Discourse-level relation extraction via graph pooling, AAAI DLG Wrokshop
- 10.1109/CVPR52688.2022.01308 · DOI · Paper page
- 10.1609/aaai.v33i01.33012580 · DOI
- 10.1109/CVPR52688.2022.00240 · DOI · Paper page
- guo, 2020, Cord19sts: Covid19 semantic textual similarity dataset, ArXiv Preprint
- guo, 2022, Multidomain learning for updating face anti-spoofing models, ECCV
Cited by
Huang, Z., Hu, J., Li, X., He, Y., Zhao, X., Peng, B., Wu, B., Huang, X., & Cheng, G. (2025) · 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52734.2025.02685 · Cited by 62
“Compared to existing IFDL (Image Forgery Detection and Localization) methods [11, 21, 38], which primarily focus on detecting tampered versus real images, SIDA is capable of handling a broader range of scenarios, including fully synthetic, tampered, and real images.”
InfluentialDiffusion Facial Forgery Detection
Cheng, H., Guo, Y., Wang, T., Nie, L., & Kankanhalli, M. (2024) · Proceedings of the 32nd ACM International Conference on Multimedia · doi:10.1145/3664647.3680797 · Cited by 47
“Subsequently, we curate approximately 20 images per identity from public resources, resulting in a pristine collection, denoted as I , which encompasses a total of 23,661 images.”
Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector
Guo, X., Song, X., Zhang, Y., Liu, X., & Liu, X. (2025) · 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52734.2025.00019 · Cited by 34
“…detection techniques, including subtle artifacts indentification [6, 22, 39, 40, 61, 90], frequency analysis [19, 44, 48, 55, 74], disentangling forgery traces via specialized neural networks [15, 20, 26, 43, 46, 78, 79, 89], modeling temporal inconsistencies [24, 62, 75, 91], among others.”
Self-citationWeakly-supervised deepfake localization in diffusion-generated images
Ţânţaru, D.-C., Oneaţă, E., & Oneaţă, D. (2024) · 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv57701.2024.00614 · Cited by 25
“For the Repaint–P2 dataset we use a subset of 9k samples to match the samples from the other approaches (see Table 3).”
BackgroundMethodEvolving from Single-modal to Multi-modal Facial Deepfake Detection: Progress and Challenges
Liu, P., Tao, Q., & Zhou, J. T. (2024) · arXiv · doi:10.48550/arxiv.2406.06965 · Cited by 6
Liu, P., Tao, Q., & Zhou, J. (2024) · Proceedings of the 1st ACM Multimedia Workshop on Multi-modal Misinformation Governance in the Era of Foundation Models · doi:10.1145/3689090.3689386 · Cited by 3
Nguyen-Le, H.-H., Tran, V.-T., Nguyen, D.-T., & Le-Khac, N.-A. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11608-4 · Cited by 28
Metadata
- Authors
- Guo, X., Liu, X., Ren, Z., Grosz, S., Masi, I., & Liu, X.
- Venue
- 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2023-06
- Pages
- 3155-3165
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W4386075856
- Cited by
- 205
- References
- 35
- FWCI
- 23.31
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.