SOTAVerified

Video Inpainting

The goal of Video Inpainting is to fill in missing regions of a given video sequence with contents that are both spatially and temporally coherent. Video Inpainting, also known as video completion, has many real-world applications such as undesired object removal and video restoration.

Source: Deep Flow-Guided Video Inpainting

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 130 papers

TitleStatusHype
ProPainter: Improving Propagation and Transformer for Video InpaintingCode5
DiffuEraser: A Diffusion Model for Video InpaintingCode4
VideoPainter: Any-length Video Inpainting and Editing with Plug-and-Play Context ControlCode4
Replace Anyone in VideosCode4
Lumiere: A Space-Time Diffusion Model for Video GenerationCode3
Advanced Video Inpainting Using Optical Flow-Guided Efficient DiffusionCode3
StereoCrafter: Diffusion-based Generation of Long and High-fidelity Stereoscopic 3D from Monocular VideosCode3
CoCoCo: Improving Text-Guided Video Inpainting for Better Consistency, Controllability and CompatibilityCode3
Towards An End-to-End Framework for Flow-Guided Video InpaintingCode3
Flow-Guided Diffusion for Video InpaintingCode2
SwapAnyone: Consistent and Realistic Video Synthesis for Swapping Any Person into Any VideoCode2
Flow-edge Guided Video CompletionCode2
Towards Unified Keyframe Propagation ModelsCode2
Exploiting Optical Flow Guidance for Transformer-Based Video InpaintingCode2
Towards Language-Driven Video Inpainting via Multimodal Large Language ModelsCode2
Flow-Guided Transformer for Video InpaintingCode2
Elevating Flow-Guided Video Inpainting with Reference GenerationCode2
Scalable Neural Video Representations with Learnable Positional FeaturesCode1
Deficiency-Aware Masked Transformer for Video InpaintingCode1
BIVDiff: A Training-Free Framework for General-Purpose Video Synthesis via Bridging Image and Video Diffusion ModelsCode1
Raformer: Redundancy-Aware Transformer for Video Wire InpaintingCode1
Bitstream-Corrupted Video Recovery: A Novel Benchmark Dataset and MethodCode1
AVID: Any-Length Video Inpainting with Diffusion ModelCode1
Fine-Grained Egocentric Hand-Object Segmentation: Dataset, Model, and ApplicationsCode1
Video Instance Shadow Detection Under the Sun and SkyCode1
VQ-NeRV: A Vector Quantized Neural Representation for VideosCode1
UMMAFormer: A Universal Multimodal-adaptive Transformer Framework for Temporal Forgery LocalizationCode1
Depth-Aware Endoscopic Video InpaintingCode1
Internal Video Inpainting by Implicit Long-range PropagationCode1
Learning Joint Spatial-Temporal Transformations for Video InpaintingCode1
Video Inpainting Localization with Contrastive LearningCode1
Inertia-Guided Flow Completion and Style Fusion for Video InpaintingCode1
Progressive Temporal Feature Alignment Network for Video InpaintingCode1
Infusion: internal diffusion for inpainting of dynamic textures and complex motionCode1
The DEVIL is in the Details: A Diagnostic Evaluation Benchmark for Video InpaintingCode1
Diverse Generation from a Single Video Made PossibleCode1
DVI: Depth Guided Video Inpainting for Autonomous DrivingCode1
Decoupled Spatial-Temporal Transformer for Video InpaintingCode1
FuseFormer: Fusing Fine-Grained Information in Transformers for Video InpaintingCode1
HNeRV: A Hybrid Neural Representation for VideosCode1
Beyond the Field-of-View: Enhancing Scene Visibility and Perception with Clip-Recurrent TransformerCode1
INR-V: A Continuous Representation Space for Video-based Generative TasksCode1
A Temporally-Aware Interpolation Network for Video Frame InpaintingCode0
CIRI: Curricular Inactivation for Residue-aware One-shot Video InpaintingCode0
Deep Video Inpainting Guided by Audio-Visual Self-SupervisionCode0
Deep Video InpaintingCode0
An Internal Learning Approach to Video InpaintingCode0
Deep Flow-Guided Video InpaintingCode0
Fast and Accurate Tensor Completion with Total Variation Regularized Tensor TrainsCode0
MotionAura: Generating High-Quality and Motion Consistent Videos using Discrete DiffusionCode0
Show:102550
← PrevPage 1 of 3Next →
All datasetsDAVISYouTube-VOS 2018HQVI (240p)HQVI (480p)HQVI (2K)YouTube-VOSHow2Sign

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1DMTPSNR33.82—Unverified
2E2FGVIPSNR33.01—Unverified
3FuseFormerPSNR32.54—Unverified
4FGVCPSNR30.8—Unverified
5STTNPSNR30.67—Unverified
6CAPPSNR30.28—Unverified
7VINetPSNR28.96—Unverified
8DFVIPSNR28.81—Unverified
9LGTSMPSNR28.57—Unverified
10FGT++LPIPS (object)0.04—Unverified
#ModelMetricClaimedVerifiedStatus
1ProPainterPSNR34.43—Unverified
2DMTPSNR34.27—Unverified
3E2FGVIPSNR33.71—Unverified
4FuseFormerPSNR33.29—Unverified
5STTNPSNR32.34—Unverified
6CAPPSNR31.58—Unverified
7LGTSMPSNR29.74—Unverified
8FGVCPSNR29.67—Unverified
9VINetPSNR29.2—Unverified
10DFVIPSNR29.16—Unverified
#ModelMetricClaimedVerifiedStatus
1STTNLPIPS0.05—Unverified
2FuseFormerLPIPS0.05—Unverified
3FGVCLPIPS0.04—Unverified
4E2FGVILPIPS0.04—Unverified
5RGVI w/o Ref.LPIPS0.04—Unverified
6ProPainterLPIPS0.04—Unverified
7RGVILPIPS0.03—Unverified
#ModelMetricClaimedVerifiedStatus
1ProPainterLPIPS0.05—Unverified
2RGVI w/o Ref.LPIPS0.04—Unverified
3FGVCLPIPS0.04—Unverified
4RGVILPIPS0.03—Unverified
#ModelMetricClaimedVerifiedStatus
1RGVI w/o Ref.LPIPS0.04—Unverified
2RGVILPIPS0.04—Unverified
#ModelMetricClaimedVerifiedStatus
1FGT++LPIPS0.03—Unverified
2FGT++*LPIPS0.02—Unverified
#ModelMetricClaimedVerifiedStatus
1INR-VL1 error4.51—Unverified