AI Paper Atlas

Research map / Architectures and efficiency / Quantization

SPA-Q: Structure-Preserving Adaptive Post-Training Quantization for Monocular Depth Estimation

Jaemin Choi, Jincheol Yang, Nahyun Lim, Yun-Seong Jeong, Matti Zinke, Hyunwoo Yu, Suk-Ju Kang

NeurIPS 2026 · Architectures and efficiency › Quantization · approach: Post-training quantization of ViTs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes SPA-Q, a structure-preserving adaptive post-training quantization method for monocular depth estimation foundation models such as Depth Anything.

Problem

Transformer-based monocular depth estimation foundation models (e.g., Depth Anything) are costly in compute and memory, and standard post-training quantization degrades them; the abstract is truncated, so the specific failure mode is only partially stated.

Approach

Quantization method named for preserving structure during adaptive post-training quantization of MDE models; the abstract is truncated so the exact mechanism is not stated.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
calibrate Q and K on pre-softmax logits instead of local per-projection error (ablation)SPA-Q · NYUv2 W4A4, DA v1 ViT-BAbsRel: 0.1908 -> 0.1839No clear effect
replace logits-level objective with attention-preserving calibration (APC) (ablation)SPA-Q · NYUv2 W4A4, DA v1 ViT-BAbsRel: 0.1839 -> 0.1444Improved
add channel-wise distribution alignment (CWDA) on top of APC (ablation)SPA-Q · NYUv2 W4A4, DA v1 ViT-BAbsRel: 0.1444 -> 0.1021Improved
add APC and CWDA vs local calibration (ablation)SPA-Q · KITTI W4A4, DA v1 ViT-BAbsRel: 0.3241 -> 0.1840Improved
use SPA-Q instead of QSCA under 4-bitDepth Anything v1 ViT-B · NYUv2delta1: 0.6845 -> 0.8991Improved
use SPA-Q instead of QSCA under 4-bitDepth Anything v1 ViT-B · KITTIAbsRel: 0.2365 -> 0.1840Improved
use SPA-Q instead of QSCA on synthetic SintelDepth Anything v1 ViT-B · SintelAbsRel: 0.4781 -> 0.3949Improved
fuse CWDA affine parameters into weights after training (ablation)SPA-Q · ViT-B efficiencymodel size and BOPs: BOPs 4493 -> 4270 GImproved
use SPA-Q instead of QSCA's auxiliary modulesDepth Anything v1 ViT-S · efficiencymodel size: 16.27 MB -> 14.16 MBImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

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