VEnhancer: Generative Space-Time Enhancement for Video Generation
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:VNZCLVYUrecord.jsonopen to challenge →
read the original abstract
We present VEnhancer, a generative space-time enhancement framework that improves the existing text-to-video results by adding more details in spatial domain and synthetic detailed motion in temporal domain. Given a generated low-quality video, our approach can increase its spatial and temporal resolution simultaneously with arbitrary up-sampling space and time scales through a unified video diffusion model. Furthermore, VEnhancer effectively removes generated spatial artifacts and temporal flickering of generated videos. To achieve this, basing on a pretrained video diffusion model, we train a video ControlNet and inject it to the diffusion model as a condition on low frame-rate and low-resolution videos. To effectively train this video ControlNet, we design space-time data augmentation as well as video-aware conditioning. Benefiting from the above designs, VEnhancer yields to be stable during training and shares an elegant end-to-end training manner. Extensive experiments show that VEnhancer surpasses existing state-of-the-art video super-resolution and space-time super-resolution methods in enhancing AI-generated videos. Moreover, with VEnhancer, exisiting open-source state-of-the-art text-to-video method, VideoCrafter-2, reaches the top one in video generation benchmark -- VBench.
This paper has not been read by Pith yet.
Forward citations
Cited by 19 Pith papers
-
PolarVSR: A Unified Framework and Benchmark for Continuous Space-Time Polarization Video Reconstruction
PolarVSR is the first unified architecture for continuous space-time polarization video reconstruction from DoFP captures, using polarization-aware implicit neural representations, a flow-guided variation loss, and a ...
-
DVG-WM: Disentangled Video Generation Enables Efficient Embodied World Model for Robotic Manipulation
DVG-WM disentangles dynamics learning and visual synthesis in video world models using flow matching and latent degradation to achieve faster inference up to 3.97 times with improved quality on LIBERO and real-world r...
-
SwiftVR: Real-Time One-Step Generative Video Restoration
SwiftVR achieves real-time generative video restoration at 1080p on consumer GPUs (26 FPS on RTX 5090) and higher resolutions on H100 via efficient dense attention and chunk-wise autoencoding.
-
DTG-Restore: Training-Free Diffusion Refinement for Generative Video Super-Resolution
Presents Decoupled Time Guidance (DTG) for training-free generative video super-resolution by temporally decoupling conditional and unconditional diffusion signals.
-
ChopGrad: Pixel-Wise Losses for Latent Video Diffusion via Truncated Backpropagation
ChopGrad truncates backpropagation to local frame windows in video diffusion models, reducing memory from linear in frame count to constant while enabling pixel-wise loss fine-tuning.
-
Taming Real-World Space-Time Video Super-Resolution with One-Step Diffusion
OSDEnhancer delivers state-of-the-art real-world space-time video super-resolution via one-step diffusion with temporal coherence and texture enrichment LoRAs plus a deformable recurrent VAE decoder.
-
Stream-DiffVSR: Low-Latency Streamable Video Super-Resolution via Auto-Regressive Diffusion
Stream-DiffVSR enables practical low-latency video super-resolution by combining a four-step distilled denoiser, auto-regressive temporal guidance, and a temporal processor in a strictly causal pipeline.
-
Beyond the Frame: Generating 360 Panoramic Videos from Perspective Videos
A generative model produces realistic and coherent 360 panoramic videos from in-the-wild perspective videos via curated online data and geometry-motion aware operations.
-
AVSR-Diff: Scale-Agnostic Diffusion Priors for Temporally Consistent Arbitrary-Scale Video Super-Resolution
AVSR-Diff uses a decoupled diffusion framework with TGFR and SAFR modules to deliver temporally consistent arbitrary-scale video super-resolution that outperforms fixed-scale and arbitrary-scale baselines.
-
LiteVSR: Lightweight Adaptation of Frozen Diffusion Transformers for Video Super-Resolution
LiteVSR performs video super-resolution on a completely frozen Diffusion Transformer via a lightweight State-Aware Adapter that uses dual-stream extraction and time-dependent cross-attention, reaching competitive qual...
-
AtlasVid: Efficient Ultra-High-Resolution Long Video Generation via Decoupled Global-Local Modeling
AtlasVid proposes a decoupled global-local diffusion framework that trains at low resolution with LoRA and generalizes to ultra-high-resolution long video synthesis via semantic proxy guidance and locality-preserving ...
-
DiffST: Spatiotemporal-Aware Diffusion for Real-World Space-Time Video Super-Resolution
DiffST delivers state-of-the-art real-world space-time video super-resolution with 17x faster inference than prior diffusion methods by using one-step sampling, cross-frame context aggregation, and video representatio...
-
Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation
PhyGenBench supplies 160 prompts across 27 physical laws and an automated LLM/VLM evaluation pipeline to measure physical commonsense compliance in current text-to-video models.
-
FADRA: Frequency-Aware Diffusion with Residual Adaptation for Video Face Restoration
FADRA restores degraded face videos by adding LQ-guided step-wise residual refinement and a frequency-aware loss to a frozen text-to-video diffusion model, achieving state-of-the-art temporal coherence and spatial fidelity.
-
DVG-WM: Disentangled Video Generation Enables Efficient Embodied World Model for Robotic Manipulation
DVG-WM disentangles dynamics learning from visual synthesis via flow matching and latent degradation to deliver faster, higher-quality video predictions for robotic manipulation.
-
TIGER: Taming Identity, Geometry, and Generative Priors for High-Quality Face Video Restoration
TIGER is a tri-prior fusion method for face video restoration using identity, geometry, and generative priors with progressive training to achieve SOTA identity fidelity and temporal stability on a new large-scale dataset.
-
Ultra Flash: Scaling Real-Time Streaming Video Generation to High Resolutions
Ultra Flash introduces a cascaded streaming super-resolution framework with specialized training, upsampling, and optimization to enable real-time high-resolution video generation from low-res diffusion models.
-
MRecover: A Conditional Generative Model for Recovering Motion-Corrupted MR images Using AI Generated Contrast
MRecover synthesizes motion-robust TSE images from T1w inputs using autoregressive slice conditioning, achieving SSIM 0.84 in-domain and volume correlations r=0.87-0.97 out-of-domain while recovering 31.8% more analyz...
-
TIGER: Taming Identity, Geometry, and Generative Priors for High-Quality Face Video Restoration
TIGER fuses identity embeddings, disentangled 3D geometry parameters, and one-step generative priors with three-stage training to achieve claimed state-of-the-art identity fidelity and temporal stability in face video...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.