Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Trajectory Consistency Distillation: Improved Latent Consistency Distillation by Semi-Linear Consistency Function with Trajectory Mapping

About

Latent Consistency Model (LCM) extends the Consistency Model to the latent space and leverages the guided consistency distillation technique to achieve impressive performance in accelerating text-to-image synthesis. However, we observed that LCM struggles to generate images with both clarity and detailed intricacy. Consequently, we introduce Trajectory Consistency Distillation (TCD), which encompasses trajectory consistency function and strategic stochastic sampling. The trajectory consistency function diminishes the parameterisation and distillation errors by broadening the scope of the self-consistency boundary condition with trajectory mapping and endowing the TCD with the ability to accurately trace the entire trajectory of the Probability Flow ODE in semi-linear form with an Exponential Integrator. Additionally, strategic stochastic sampling provides explicit control of stochastic and circumvents the accumulated errors inherent in multi-step consistency sampling. Experiments demonstrate that TCD not only significantly enhances image quality at low NFEs but also yields more detailed results compared to the teacher model at high NFEs.

Jianbin Zheng, Minghui Hu, Zhongyi Fan, Chaoyue Wang, Changxing Ding, Dacheng Tao, Tat-Jen Cham• 2024

Related benchmarks

TaskDatasetResultRank
Aesthetic EvaluationCC3M SDXL 1.0 (test)
HPS0.2576
27
Image GenerationCC3M SDXL v1.0 (test)
FID46.4
27
Text-to-Image GenerationCOCO 5k SDXL 1024px 2017 (val)
CLIP Score32.45
8
Layout ControlCOCO
AP4.8
7
Seed Identifiability ClassificationMS-COCO
Self-Identifiability Accuracy87.3
7
Text-to-Image GenerationCOCO 5k SD15 512px 2017 (val)
CLIP Score30.62
5
Showing 6 of 6 rows

Other info

Follow for update