A Low-bit Quantization and Pose Evaluation of a Video Diffusion Transformer against the Noise Floor

Kinematic acceptance tests for post-training quantization of human-video diffusion transformers assume that low-bit arithmetic degrades generated motion.This work tests the assumption on DynamiCtrl at W8A8 and W4A8, with refinement of the driving pose and of the extracted skeleton.Every comparison is referred to the difference between two bfloat16 renders that differ only in the seed, with inference at the clip level.No difference from the bfloat16 reference was detected in any extracted skeleton statistic, whereas pose error and pixel distance were smaller than between seeds; equivalence within half of the seed-to-seed change was not established.Calibrating the quantizer on other subjects left pose and pixel error unchanged.With fixed thresholds the kinematic gate rejected 11 of 12 seed pairs; with calibrated thresholds, 21 of 24 W8A8 and 20 of 24 W4A8 videos passed, against 9 of 12 seed pairs, rates that depend on the number of calibration clips.

Authors

Institutions

Publication Details

Journal
International journal of intelligent engineering and systems
Published
2026-11-17
DOI
https://doi.org/10.22266/ijies2026.1231.64
Primary Topic
Generative Adversarial Networks and Image Synthesis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A Low-bit Quantization and Pose Evaluation of a Video Diffusion Transformer against the Noise Floor

Ahmad Reza Naghsh‐Nilchi, Alaa Abdullah Albaghdadi
International journal of intelligent engineering and systems
Generative Adversarial Networks and Image Synthesis
article

A Low-bit Quantization and Pose Evaluation of a Video Diffusion Transformer against the Noise Floor

Ahmad Reza Naghsh‐Nilchi, Alaa Abdullah Albaghdadi
article en

Abstract

Kinematic acceptance tests for post-training quantization of human-video diffusion transformers assume that low-bit arithmetic degrades generated motion.This work tests the assumption on DynamiCtrl at W8A8 and W4A8, with refinement of the driving pose and of the extracted skeleton.Every comparison is referred to the difference between two bfloat16 renders that differ only in the seed, with inference at the clip level.No difference from the bfloat16 reference was detected in any extracted skeleton statistic, whereas pose error and pixel distance were smaller than between seeds; equivalence within half of the seed-to-seed change was not established.Calibrating the quantizer on other subjects left pose and pixel error unchanged.With fixed thresholds the kinematic gate rejected 11 of 12 seed pairs; with calibrated thresholds, 21 of 24 W8A8 and 20 of 24 W4A8 videos passed, against 9 of 12 seed pairs, rates that depend on the number of calibration clips.

International journal of intelligent engineering and systemsVol. 19(12)
University of Isfahan (IR)
Generative Adversarial Networks and Image Synthesis
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon