AI RESEARCH
Latent Inter-Frame Pruning: A Training-Free Method Bridging Traditional Video Compression and Modern Diffusion Transformers for Efficient Generation
arXiv CS.CV
•
ArXi:2604.23858v1 Announce Type: new Video generation, while capable of generating realistic videos, is computationally expensive and slow, prohibiting real-time applications. In this paper, we observe that video latents encoded via an autoencoder under the Latent Diffusion Model (LDM) framework contain redundancy along the temporal axis.