Diffusion Transformer Video Generation in Compressed Latent Space

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional systems struggle to generate high-resolution and high-frame-rate videos efficiently, often requiring cascaded diffusion models that suffer from limited detail and quality due to low-resolution generation, and mismatched training and test distributions.

Innovation Solution

A single latent diffusion model processes video frames through latent representations, downsampling to a compressed latent space, and leveraging neural network blocks on a shorter sequence of tokens to generate high-quality videos with reduced computational resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a cascaded diffusion model is used to generate high resolution videos, then spatial resolution can be improved, but manufacturing precision deteriorates due to limited detail generation capability

Engineering Contradiction:
Improvespatial resolutionVSAvoiddetail generation quality
Core Design Contradiction:
Measurement precisionVSManufacturing precision

Solution Approach 1:

The video generation process is segmented into two distinct stages: a first diffusion model generates lower resolution video frames to establish temporal coherence, while a second diffusion model generates higher resolution video frames to achieve detailed quality. This segmentation allows each model to specialize in different resolution requirements, resolving the contradiction between spatial resolution and detail generation quality.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If a cascaded diffusion model is used to generate high frame rate videos, then temporal resolution can be improved, but device complexity increases

Engineering Contradiction:
Improvetemporal resolutionVSAvoidmodel complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The temporal resolution improvement is achieved by segmenting the generation process into two models with different temporal characteristics. The first model operates at lower temporal resolution to establish base temporal coherence, while the second model enhances temporal resolution for final output. This segmentation avoids the complexity of a single cascaded model while maintaining high frame rate capability.

Inventive Principle:
Principle #1Segmentation

3Productivity

If extensive compression is applied to reduce computational resources, then productivity can be improved, but loss of information increases

Engineering Contradiction:
Improvecomputational efficiencyVSAvoidvideo quality
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The computational process is segmented into two passes: the first diffusion model processes compressed representations to establish temporal coherence efficiently, while the second diffusion model processes the full-resolution data to preserve video quality. This segmentation allows aggressive compression during the computationally intensive first pass while maintaining quality in the final output pass.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4660879A1Generating temporal sequences using diffusion transformer neural networks
Publication Date: 2025.12.10 GOOGLE LLC
  • EP4660879A1 patent drawingFigure 1A
  • EP4660879A1 patent drawingFigure 1B
  • EP4660879A1 patent drawingFigure 2

AI summary

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a temporal sequence of data elements conditioned on an input. One of the methods includes obtaining the input, wherein the input comprises a noise input comprising a plurality of latent representations for the output temporal sequence; updating each latent representation using a latent denoising neural network; and generating the output temporal sequence of data elements by processing the updated latent representations using a decoder neural network.