AI Paper Atlas

Research map / Architectures and efficiency / Quantization

POCA: Post-training Quantization with Temporal Alignment for Codec Avatars

Jian Meng, Yuecheng Li, Leo (Chenghui) Li, Syed Shakib Sarwar, Dilin Wang, Jae-sun Seo

ECCV 2024 · Architectures and efficiency › Quantization · approach: Post-training quantization of ViTs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add view-aware visibility masking and shape-wise dynamic filtering to PTQAvatar texture decoder, W8A8 · Multiface avatar 2643814 frontFovVideoVDP: 5.81 -> 6.19Improved
apply POCA instead of standard PTQ at 6-bitAvatar texture decoder, W6A6 · Multiface avatar 2643814 frontFovVideoVDP: 4.50 -> 6.06Improved
replace 8-bit QAT with POCA PTQAvatar decoder quantization · Multiface avatar 2643814 front, W8A8FovVideoVDP: 5.92 -> 6.19Improved
use PTQ with POCA instead of QATAvatar decoder quantization · Quantization timequantization time: >10 hrs -> ~13 minImproved
raise SWD threshold above 0.9 (ablation)POCA · Avatar 2643814FovVideoVDP / SSIM: FovVideoVDP 6.213 -> 6.185 (tau 0.95), 6.121 (tau 0.975)Hurt
use sparse SWD calibration (80% sparsity) on super-resolutionSRResNet W8A8 · Set5/Set14/BSD100/Urban100 x4PSNR: BSD100 27.618 (SR-PTQ) -> 27.620; Set5 32.207 -> 31.898Mixed
apply 8-bit standard PTQ to avatar decoderAvatar decoder · Rendered avatar in motiontemporal noise: pixel shift +-0.44, +-0.47, +-0.25 on R, G, BHurt

Later papers that compared with this method

1 later accepted paper reports POCA as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy