AI Paper Atlas

Research map / 3D vision / Face and shape

SeqXY2SeqZ: Structure Learning for 3D Shapes by Sequentially Predicting 1D Occupancy Segments From 2D Coordinates

Zhizhong Han, Guanhui Qiao, Yu-Shen Liu, Matthias Zwicker

ECCV 2020 · 3D vision › Face and shape · approach: Primitive-based shape abstraction · 33 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 attention (ablation)SeqXY2SeqZ · ShapeNet chair single-image reconstructionIoU: 47.5 -> 50.3Improved
use GRU cells instead of LSTM (ablation)SeqXY2SeqZ · ShapeNet chair single-image reconstructionIoU: 49.8 -> 50.3Improved
use bidirectional RNN encoder instead of single direction (ablation)SeqXY2SeqZ · ShapeNet chair single-image reconstructionIoU: 48.8 -> 50.3Improved
share one location embedding matrix across X, Y and Z (ablation)SeqXY2SeqZ · ShapeNet chair single-image reconstructionIoU: 50.3 -> 49.1No clear effect
tubelize along Z instead of Y (ablation)SeqXY2SeqZ · ShapeNet airplane auto-encodingIoU: 90.35 -> 89.96No clear effect
tubelize along X instead of Y (ablation)SeqXY2SeqZ · ShapeNet airplane auto-encodingIoU: 90.35 -> 90.21No clear effect
increase embedding and hidden size to 1024-2048SeqXY2SeqZ · ShapeNet chair auto-encodingIoU: 74.32 -> 93.10Improved
use 2D-function tube prediction instead of 3D implicit function (DISN)SeqXY2SeqZ · ShapeNet single-image 32^3mean IoU: 57.0 -> 63.6Improved
use 2D-function tube prediction instead of DIB-RSeqXY2SeqZ · ShapeNet single-image 32^3mean IoU: 61.2 -> 63.6Improved
use 2D-function tube prediction instead of DISNSeqXY2SeqZ · Single-image reconstruction costCPU time: 14.68s -> 8.79sImproved
use 2D-function tube prediction instead of OccNetSeqXY2SeqZ · Single-image reconstruction costmemory: 1175 MB -> 286MBImproved

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