REVIEW 7 cited by
DifFace: Blind Face Restoration with Diffused Error Contraction
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
abstract
While deep learning-based methods for blind face restoration have achieved unprecedented success, they still suffer from two major limitations. First, most of them deteriorate when facing complex degradations out of their training data. Second, these methods require multiple constraints, e.g., fidelity, perceptual, and adversarial losses, which require laborious hyper-parameter tuning to stabilize and balance their influences. In this work, we propose a novel method named DifFace that is capable of coping with unseen and complex degradations more gracefully without complicated loss designs. The key of our method is to establish a posterior distribution from the observed low-quality (LQ) image to its high-quality (HQ) counterpart. In particular, we design a transition distribution from the LQ image to the intermediate state of a pre-trained diffusion model and then gradually transmit from this intermediate state to the HQ target by recursively applying a pre-trained diffusion model. The transition distribution only relies on a restoration backbone that is trained with $L_2$ loss on some synthetic data, which favorably avoids the cumbersome training process in existing methods. Moreover, the transition distribution can contract the error of the restoration backbone and thus makes our method more robust to unknown degradations. Comprehensive experiments show that DifFace is superior to current state-of-the-art methods, especially in cases with severe degradations. Code and model are available at https://github.com/zsyOAOA/DifFace.
Forward citations
Cited by 7 Pith papers
-
Robust ID-Specific Face Restoration via Alignment Learning
RIDFR injects a reference person's identity into diffusion-based face restoration and uses Alignment Learning across multiple same-identity references to suppress pose, expression, and makeup interference.
-
InstaRevive: One-Step Image Enhancement via Dynamic Score Matching
A one-step image enhancer that combines dynamically controlled score-based distillation with caption prompts, matching multi-step diffusion quality on face and super-resolution benchmarks.
-
INDIGO+: A Unified INN-Guided Probabilistic Diffusion Algorithm for Blind and Non-Blind Image Restoration
An invertible neural network trained to mimic image degradations is used to steer a pretrained diffusion model in every sampling step, giving a blind and non-blind image restoration algorithm.
-
Proxies for Distortion and Consistency with Applications for Real-World Image Restoration
The paper introduces degradation-estimation-based proxies for MSE, LPIPS, and consistency so that real-world image restoration methods can be ranked without ground truth.
-
SeedVR: Seeding Infinity in Diffusion Transformer Towards Generic Video Restoration
A 2.48B-parameter diffusion transformer with shifted-window attention and a causal video autoencoder reports competitive perceptual-quality video restoration across synthetic, real-world, and AI-generated benchmarks.
-
F-Bench: Rethinking Human Preference Evaluation Metrics for Benchmarking Face Generation, Customization, and Restoration
FaceQ, a new 12K-image benchmark with multi-dimensional human preference scores, reveals that existing quality metrics poorly match human judgment on AI-generated faces, and F-Eval, an instruction-tuned LMM, outperforms them.
-
Beyond Pixels: Text Enhances Generalization in Real-World Image Restoration
A restoration-specific captioner that adaptively generates detailed text descriptions improves the generalization of text-to-image diffusion models on real-world image restoration.
Discussion (0). Continue with ORCID to comment.