Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

T2V-Turbo-v2: Enhancing Video Generation Model Post-Training through Data, Reward, and Conditional Guidance Design

About

In this paper, we focus on enhancing a diffusion-based text-to-video (T2V) model during the post-training phase by distilling a highly capable consistency model from a pretrained T2V model. Our proposed method, T2V-Turbo-v2, introduces a significant advancement by integrating various supervision signals, including high-quality training data, reward model feedback, and conditional guidance, into the consistency distillation process. Through comprehensive ablation studies, we highlight the crucial importance of tailoring datasets to specific learning objectives and the effectiveness of learning from diverse reward models for enhancing both the visual quality and text-video alignment. Additionally, we highlight the vast design space of conditional guidance strategies, which centers on designing an effective energy function to augment the teacher ODE solver. We demonstrate the potential of this approach by extracting motion guidance from the training datasets and incorporating it into the ODE solver, showcasing its effectiveness in improving the motion quality of the generated videos with the improved motion-related metrics from VBench and T2V-CompBench. Empirically, our T2V-Turbo-v2 establishes a new state-of-the-art result on VBench, with a Total score of 85.13, surpassing proprietary systems such as Gen-3 and Kling.

Jiachen Li, Qian Long, Jian Zheng, Xiaofeng Gao, Robinson Piramuthu, Wenhu Chen, William Yang Wang• 2024

Related benchmarks

TaskDatasetResultRank
Text-to-Video GenerationVBench
Quality Score85.93
111
Video GenerationVBench
Quality Score83.93
102
Text-to-Video GenerationVBench standard 1.0 (test)
Subject Consistency95.5
7
Text-to-Video GenerationNeuS-V 360-prompt set 1.0 (full set)
Nature Score0.564
4
Text-to-Video GenerationNeuS-V 160-prompt human-evaluated subset 1.0
PCC (Nature)0.46
4
Showing 5 of 5 rows

Other info

Follow for update