Organizations: College of Physics and Information Engineering, Guangdong University of Education, China · Cloudbutterfly Technology · School of Electronic Information, Huzhou College, Huzhou, 313000, China
Paired unretouched/retouched face videos are costly to produce because professional editing must remain consistent over time, while releasing identifiable face footage raises privacy, consent, and governance concerns. We present DeepVFR, an initial controlled synthetic benchmark-not a new retouching architecture-for paired face video retouching. Starting from paired synthetic DeepFRR images, Wan2.2 generates mild-motion source clips, and backend-specific face transfer constructs synchronized proxy retouching targets, followed by automatic validity screening and manual visual inspection. DeepVFR contains 1,024 pairs of 150-frame clips at 30 fps, evaluated under an identity-disjoint protocol with explicit frame correspondence; release metadata will record the construction route. Within-benchmark baselines, external checkpoints applied to DeepVFR, qualitative analysis, and a 17-participant ranking study establish reference behavior in the controlled domain. The constructed targets exhibit a substantial low-preference tail in the human study. Because both streams inherit synthetic content and the targets additionally inherit commercial-API or face-transfer artifacts, the benchmark neither replaces artist-retouched captured video nor establishes synthetic-to-real generalization. DeepVFR provides a starting point for paired supervision and controlled evaluation. Data, individual study rankings, and training logs will be made publicly available upon acceptance. Broader motion coverage and real-video validation remain future work.
Figures & tables
Benchmark
Scale
Temporal unit
Motion source
Public split
Paired beauty task
FFHQR-Seq ( Xue et al., 2024b )
56k/7k/7k seq.
5 frames/sample
Geometric augmentation
Yes
Yes
MRFV ( Xue et al., 2024b )
200 videos
≥ 500 frames/video
Real portrait videos
No
Yes
DeepVFR (ours)
1,024 videos
150 frames/video
Paired I2V videos
Planned
Yes
Table 1: Representative supervision sources for video face retouching.
Dataset
Scale
Paired
Main focus
MIT-Adobe FiveK ( Bychkovsky et al., 2011 )
5k images
Yes
General photo retouching
PPR10K ( Liang et al., 2021 )
11,161 portraits
Yes
Portrait photo retouching
SCUT-FBP5500 ( Liang et al., 2018 )
5,500 faces
No
Beauty perception
EFrR ( Lin et al., 2019 )
Small exemplar set
Yes
Freckle/skin retouching
FFHQ/FFHQR ( Karras et al., 2019 ; Shafaei et al., 2021 )
Face images
Partial
Face generation/retouching
RetouchingFFHQ ( Ying et al., 2023 )
> 0.5M images
Conditional
Retouching detection
Table 2: Representative image-level resources related to facial beauty and retouching. “Paired” indicates whether an original/beauty or before/after image relation is available.
Figure 1: Overview of the DeepVFR construction pipeline. An I2V model animates the raw member of each DeepFRR pair; backend-specific transfer and parsing-guided blending then construct a frame-synchronized proxy target from the paired retouched image. The encoder/ID-block/decoder schematic illustrates the complementary SimSwap route; it does not describe the proprietary Akool backend. The right side shows representative raw/target frames.
Figure 2: Representative DeepVFR samples. Each group contains three frames from one clip; the six examples were selected from metadata strata crossing age group (child, youth, elderly) and gender (male, female).
Model
PSNR ↑
SSIM ↑
LPIPS ↓
LPIPS-T
RGBFlicker
RestoreFormer
24.0368
0.7023
0.2383
0.010897
0.001062
GPEN
23.9788
0.7299
0.2283
0.010896
0.000828
RetouchFormer
27.9078
0.8529
0.0937
0.005648
0.000810
RetouchGPT
33.7426
0.9505
0.0335
0.005435
0.000816
VRetouchEr
33.1804
0.9521
0.0381
0.003507
0.000642
Table 3: Results on the official identity-disjoint DeepVFR test split. Bold values mark the best target-referenced fidelity scores. Temporal columns describe unaligned adjacent-frame change; smaller values do not establish better temporal quality.
Model
Source supervision
PSNR ↑
SSIM ↑
LPIPS ↓
RetouchFormer
FFHQR image pairs
26.5413
0.8087
0.1405
RetouchGPT
FFHQR + instructions
26.6554
0.8120
0.1374
VRetouchEr
FFHQR-Seq pseudo-sequences
35.8251
0.9469
0.0451
Table 4: External checkpoints applied to all 1,024 DeepVFR clips (153,600 frames). “Source supervision” follows the corresponding method paper; the comparison is not fairness-controlled.
Face video restoration from degraded observations is challenging, as it requires simultaneously recovering visual fidelity, temporal consistency, and subject identity. Existing approaches are often either reference-free, which can lead to identity loss when person-specific facial details are lost, or subject-specific, which limits generalization to unseen identities. We propose a subject-agnostic, reference-guided framework for identity-preserving face video restoration. Our method introduces bimodal perceptual-descriptive identity conditioning into a pretrained flow-based text-to-video generator and employs a two-stage training strategy to strengthen identity guidance during restoration. Experiments show that our approach improves restoration fidelity, temporal consistency, and identity preservation, achieving superior performance under challenging video degradations, including downsampling, blur, noise, and compression artifacts. The code is available under: https://github.com/batuhanntosun/RG-FVR.
Cem Eteke, Batuhan Tosun, Eckehard Steinbach
Chair of Media Technology Munich Institute of Robotics and Machine Intelligence School of Computation, Information, and Technology Technical University of Munich 80333 Munich, Germany
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.
Video face swapping has no natural paired supervision: no real footage exists of one person's face performing another person's video. The strongest current answer, DreamID-V's SyncID-Pipe, mints pairs by replacing the identity in exactly two frames of a real clip -- the first and the last -- and regenerating the rest from a pose sequence alone. Pose carries no appearance evidence of the swapped-in identity, so over long clips, occlusions, and extreme pose excursions the synthesized identity has a long unanchored span on which to drift; no published ablation examines anchor count or placement. We propose Adaptive Identity Anchoring (AIA): (i) generalize the synthesizer to arbitrary anchor sets, architecturally natural for diffusion-forcing-style transformers where conditioning on a frame is clamping its tokens to zero noise; (ii) place anchors by a closed feedback loop that scores every generated frame against the real reference identity and inserts an image-face-swapped anchor at the worst-scoring frame until the pair passes a threshold or exhausts a budget; (iii) reuse the loop's verdict as an automatic data filter. A second pathology, the beauty-filter look of over-smoothed skin, has the same root cause: micro-texture, like identity, is priced by none of the pipeline's objectives. We therefore pair AIA with Reality-Referenced Texture Restoration: matched re-graining from each real frame's non-face regions, band-split transfer of sub-identity micro-texture from the real footage, and a second, spectral acceptance channel refereed by the footage's own spectrum. Identity-anchor density, we argue, is a controllable quality dial, and we specify falsifiable experiments -- drift-versus-gap curves, uniform-versus-adaptive placement at matched budgets, student training on AIA-minted data, and texture ablations with a human beauty-filter study -- that would validate or refute the proposal.