HonestFace: Towards Honest Face Restoration with One-Step Diffusion Model

Jingkai Wang; Wu Miao; Jue Gong; Zheng Chen; Xing Liu; Hong Gu; Yutong Liu; Yulun Zhang

cs.CVMay 24, 2025

HonestFace: Towards Honest Face Restoration with One-Step Diffusion Model

Authors: Jingkai Wang, Wu Miao, Jue Gong, Zheng Chen, Xing Liu, Hong Gu, Yutong Liu, Yulun Zhang

Organizations: School of Computer Science, Shanghai Jiao Tong University, Shanghai, China · vivo BlueImage Lab, vivo Mobile Communication Co., Ltd., Hangzhou, China

Abstract

Face restoration has achieved significant advancements through the years of development. However, maintaining high fidelity and authenticity while avoiding artifacts remains challenging, especially in extreme degradation scenarios. This highlights the need for models that are more ``honest'' in their reconstruction from low-quality inputs, accurately reflecting original characteristics. In this work, we propose HonestFace, a novel approach for honest face restoration with identity consistency and realistic textures. To achieve this, HonestFace incorporates several key components. First, we propose an identity embedder to effectively capture and preserve crucial identity features from both the low-quality input and multiple reference faces. Second, a masked face alignment method is presented to enhance fine-grained details and textural authenticity. Furthermore, we present a new real-world reference-based face restoration dataset, MultiRefCeleb-Test, with multiple high-quality references and low-quality inputs. Leveraging these contributions within a one-step diffusion model framework, HonestFace delivers excellent restoration results in terms of facial fidelity and realism. Experiments demonstrate that our approach surpasses state-of-the-art methods, achieving superior performance in both visual quality and quantitative assessments. The code and pre-trained models are available at https://github.com/jkwang28/HonestFace .

Figures & tables

Figure 1. Real-world concert scenes captured by smartphones. From left to right: Buyi Mao, Adele, Jason Zhang, Bibi Zhou, Fish Leong, and G.E.M. Concerts are typically held in the evening, and photos are predominantly taken using telephoto lenses. For celebrities like these, who have abundant high-definition images available online, smartphone manufacturers can directly embed high-resolution ID information, enabling automatic matching and background processing when users take photos.

Panel of Figure 2 — Figure 2. Comparisons of face restoration methods. HonestFace restores a more natural, faithful and ID-consistent face.

Figure 3. Overall training pipeline of HonestFace. First, the LQ input $x_{L}$ is encoded into $z_{L}$ by the VAE encoder $E_{\phi}$ . Meanwhile, $x_{L}$ and HQ references $R=\{r_{i}\}$ pass through IDE and VRE, then fused into the prompt embedding $p$ . Next, the UNet predicts $\varepsilon_{\theta}$ to estimate $\hat{z}_{H}$ . Finally, the VAE decoder $D_{\phi}$ reconstructs the output $\hat{x}_{H}$ . Generator and discriminator are trained alternately.

Panel of Figure 4 — Figure 4. Key modules in HonestFace pipeline. ( left , a) Facial feature extractor: Generates a feature embedding $p_{F}$ from the most geometrically similar eye patches, selected from the optimal references; ( middle , b) Face identity encoder: Encodes multiple reference faces to produce a combined identity representation $p_{I}$ ; ( right , c) Masked face alignment: A novel perceptual loss that strategically focuses the restoration on perceptually crucial areas, ensuring the fidelity of key textures and natural colors.

Figure 5. Overview of the MultiRefCeleb-Test benchmark. Each identity features multiple high-quality reference images alongside several low-quality inputs. The dataset spans diverse ethnicities and age ranges. It incorporates a wide spectrum of authentic real-world degradations and challenging cases, such as occlusions, varied illumination, and diverse facial expressions.

Panel of Figure 6 — Figure 6. Visual comparison of synthetic datasets CelebHQRef-Test and Reface-Test. Please zoom in for a better view.

Datasets	Methods	LPIPS $\downarrow$	DISTS $\downarrow$	C-IQA $\uparrow$	M-IQA $\uparrow$	MUSIQ $\uparrow$	FID $\downarrow$	Deg. $\downarrow$	LMD $\downarrow$
CelebHQRef-Test	ASFFNet ( Li et al., 2020b )	0.1969	0.1411	0.6067	0.5346	71.181	45.618	35.636	3.6047
	DMDNet ( Li et al., 2022 )	0.2191	0.1499	0.6417	0.5165	71.119	52.406	34.258	2.6437
	PGDiff ( Yang et al., 2023 )	0.2674	0.1559	0.5735	0.4953	70.762	60.693	46.248	3.6478
	FaceMe ( Liu et al., 2025 )	0.1945	0.1412	0.6606	0.5758	71.598	46.423	30.578	2.0107
	MGFR ( Tao et al., 2025 )	0.2311	0.1520	0.6224	0.6475	70.626	52.103	38.322	2.5274
	HonestFace (ours)	0.1809	0.1246	0.6673	0.6858	71.996	41.021	26.106	1.7798

Table 1. Comparison with the reference-based methods. C-IQA stands for CLIP-IQA, and M-IQA stands for MANIQA.

Panel of Figure 7 — Figure 7. Visual comparison on the real-world MultiRefCeleb-Test dataset. Please zoom in for a better view.

	CelebHQRef-Test						Reface-Test						MultiRefCeleb-Test
Methods	PSNR $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$	DISTS $\downarrow$	M-IQA $\uparrow$	FID $\downarrow$	PSNR $\uparrow$	SSIM $\uparrow$	LPIPS $\downarrow$	DISTS $\downarrow$	M-IQA $\uparrow$	FID $\downarrow$	M-IQA $\uparrow$	FID $\downarrow$
CodeFormer ( Zhou et al., 2022 )	25.49	0.7100	0.1791	0.1419	0.5827	48.37	25.65	0.7082	0.1872	0.1497	0.5699	33.57	0.5483	60.61
DAEFR ( Tsai et al., 2024 )	22.85	0.6549	0.2058	0.1439	0.6019	49.35	23.07	0.6574	0.2092	0.1498	0.5842	34.10	0.5541	48.52
OSEDiff ( Wu et al., 2024a )	24.07	0.7105	0.2077	0.1533	0.5229	57.06	24.21	0.7110	0.2076	0.1553	0.5122	46.31	0.7057	84.85
OSDFace ( Wang et al., 2025 )	23.51	0.6742	0.2050	0.1373	0.6316	46.51	23.54	0.6718	0.2156	0.1481	0.6257	32.59	0.7148	63.11
HonestFace (ours)	25.17	0.7147	0.1809	0.1246	0.6858	41.02	25.25	0.7118	0.1784	0.1235	0.6717	25.75	0.6649	57.32

Table 2. Quantitative comparison with the no-reference methods. M-IQA stands for MANIQA.

Table 3. Comprehensive quantitative evaluations. (a) Quantitative comparisons with state-of-the-art reference-based methods on the real-world dataset. (b-e) More experiments for color shift evaluation, ablation studies, and inference complexity comparison. All ablation study experiments are conducted on CelebHQRef-Test. C-IQA stands for CLIP-IQA and M-IQA stands for MANIQA.

Image Restoration

Image restoration uses computational methods to recover useful visual content from noisy, blurred, low-resolution, or otherwise degraded images.

See all 372 papers

DVFace: Spatio-Temporal Dual-Prior Diffusion for Video Face Restoration
Apr 16, 2026Zheng Chen, Bowen Chai, Rongjun Gao +7Video Diffusion Models Image Restoration
Shanghai Jiaotong University · Meituan Inc
HDRFace: Rethinking Face Restoration with High-Dimensional Representation
May 14, 2026Zirui Wang, Xianhui Lin, Yi Dong +7Diffusion Models for Image Restoration Diffusion Model Guidance
City University of Hong Kong · vivo BlueImage Lab, vivo Mobile Communication Co., Ltd
IConFace: Identity-Structure Asymmetric Conditioning for Unified Reference-Aware Face Restoration
May 4, 2026Axi Niu, Jinyang Zhang, Senyan QingImage Restoration
Northwestern Polytechnical University

Apr 16, 2026J/K move · Enter open · S save

DVFace: Spatio-Temporal Dual-Prior Diffusion for Video Face Restoration

Zheng Chen, Bowen Chai, Rongjun Gao, Mingtao Nie+6

Shanghai Jiaotong University · Meituan Inc

Video Diffusion Models Image Restoration Video Restoration Diffusion Models

Open paper AskPDF arXiv

Video face restoration aims to enhance degraded face videos into high-quality results with realistic facial details, stable identity, and temporal coherence. Recent diffusion-based methods have brought strong generative priors to restoration and enabled more realistic detail synthesis. However, existing approaches for face videos still rely heavily on generic diffusion priors and multi-step sampling, which limit both facial adaptation and inference efficiency. These limitations motivate the use of one-step diffusion for video face restoration, yet achieving faithful facial recovery alongside temporally stable outputs remains challenging. In this paper, we propose, DVFace, a one-step diffusion framework for real-world video face restoration. Specifically, we introduce a spatio-temporal dual-codebook design to extract complementary spatial and temporal facial priors from degraded videos. We further propose an asymmetric spatio-temporal fusion module to inject these priors into the diffusion backbone according to their distinct roles. Evaluation on various benchmarks shows that DVFace delivers superior restoration quality, temporal consistency, and identity preservation compared to recent methods. Code: https://github.com/zhengchen1999/DVFace.

Abstract

Image Restoration

Explore similar work

DVFace: Spatio-Temporal Dual-Prior Diffusion for Video Face Restoration

HDRFace: Rethinking Face Restoration with High-Dimensional Representation

IConFace: Identity-Structure Asymmetric Conditioning for Unified Reference-Aware Face Restoration