AI Paper Atlas

Research map / Point clouds / Point cloud classification

Neural TMDlayer: Modeling Instantaneous Flow of Features via SDE Generators

Zihang Meng, Vikas Singh, Sathya N. Ravi

ICCV 2021 · Point clouds › Point cloud classification · approach: Equivariant point cloud networks · 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
use TMDlayer at inference time instead of dropping it (m=1) (ablation)ResNet-18 + TMDlayer · CIFAR10 (ResNet-18)accuracy: 75.15 -> 87.35Improved
increase test batch size m used to build L_m from 1 to 100 (ablation)ResNet-18 + TMDlayer · CIFAR10 (ResNet-18)accuracy: 87.35 -> 88.52Improved
increase test batch size m from 150 to 200 (ablation)ResNet-18 + TMDlayer · CIFAR10 (ResNet-18)accuracy: 88.55 -> 88.25No clear effect
add TMDlayer to ResNet-18ResNet-18 · CIFAR10 (ResNet-18)accuracy: 88.27 -> 88.52No clear effect
add TMDlayer to ResNet-18 under input noise sigma=0.03ResNet-18 · CIFAR10 with input noiseaccuracy: 75.85 -> 77.96Improved
add TMDlayer to ResNet-18 under input noise sigma=0.05ResNet-18 · CIFAR10 with input noiseaccuracy: 53.87 -> 56.18Improved
replace PCT transformer layer with TMDlayer on a feed-forward layerPCT · ModelNet40 classificationoverall accuracy: 93.2 -> 93.0No clear effect
add Chan-Vese update module to the CNN baseline (ablation)CNN baseline · Vaihingen buildings segmentationaverage IoU: 68.9 -> 73.5Improved
add TMDlayer on top of deep active contour model (ablation)deep active contour model · Vaihingen buildings segmentationaverage IoU: 73.5 -> 74.6Improved
add TMDlayer to EGNN (non-transductive)EGNN · miniImageNet 5-way 5-shotaccuracy: 66.85 -> 68.35Improved
add TMDlayer to EGNN with transductionEGNN · miniImageNet 5-way 5-shotaccuracy: 76.37 -> 77.78Improved
add TMDlayer to EGNN-semi with 20% labelsEGNN-semi · miniImageNet semi-supervisedaccuracy: 61.88 -> 63.14Improved

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