Video Generation On Ucf 101

评估指标

FVD16
Inception Score
KVD16

评测结果

各个模型在此基准测试上的表现结果

Paper TitleRepository
MCVD2460-148Latent Video Diffusion Models for High-Fidelity Long Video Generation
VDM1396-116Latent Video Diffusion Models for High-Fidelity Long Video Generation
TGAN-v2 (128x128)1209--Latent Video Diffusion Models for High-Fidelity Long Video Generation
MCVD (64x64)1143--MCVD: Masked Conditional Video Diffusion for Prediction, Generation, and Interpolation
MoCoGAN-HD (256x256, unconditional)70033.95-A Good Image Generator Is What You Need for High-Resolution Video Synthesis
MagicVideo (256x256, text-conditional)699--MagicVideo: Efficient Video Generation With Latent Diffusion Models-
TATS (256x256)635-55Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer
DIGAN (128x128, unconditional)57732.70-Generating Videos with Dynamics-aware Implicit Generative Adversarial Networks
LVDM (256x256, unconditional)552-42Latent Video Diffusion Models for High-Fidelity Long Video Generation
Video LDM (320x512, text-conditional)550.6133.45-Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models
LAVIE (320x512, text-conditional)526.30--LAVIE: High-Quality Video Generation with Cascaded Latent Diffusion Models
DIGAN (128x128, class-conditional)46559.6839.6Generating Videos with Dynamics-aware Implicit Generative Adversarial Networks
MeBT (128x128, unconditional)43865.93-Towards End-to-End Generative Modeling of Long Videos with Memory-Efficient Bidirectional Transformers
TATS (128x128, unconditional)42057.63-Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer
MMVG (128x128, unconditional)39558.3-Tell Me What Happened: Unifying Text-guided Video Completion via Multimodal Masked Video Generation
LVDM (256x256, unconditional)372-27Latent Video Diffusion Models for High-Fidelity Long Video Generation
Make-A-Video (Zero-shot, 256x256, class-conditional)367.2333-Make-A-Video: Text-to-Video Generation without Text-Video Data
PYoCo (Zero-shot, 64x64, text-conditional)355.1947.76-Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models-
VideoPoet (text-conditional)35538.44-VideoPoet: A Large Language Model for Zero-Shot Video Generation-
VideoAssembler (Zero-shot, 256x256, class-conditional)346.8448.01-MagDiff: Multi-Alignment Diffusion for High-Fidelity Video Generation and Editing
0 of 46 row(s) selected.
Video Generation On Ucf 101 | SOTA | HyperAI超神经