AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Deep Supervision with Intermediate Concepts

Chi Li, M. Zeeshan Zia, Quoc-Huy Tran, Xiang Yu, Gregory D. Hager, Manmohan Krishna Chandraker

TPAMI 2018 · Architectures and efficiency › CNN design · approach: Filter design and trained-filter analysis · 87 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
supervise hidden layers with ordered intermediate concepts instead of plain single-task output (ablation)DISCO · CIFAR100classification error (%): 23.31 -> 22.46Improved
order concepts by structure instead of random concept assignment (ablation)DISCO · CIFAR100classification error (%): 27.53 -> 22.46Improved
supervise 3D shape concepts in intermediate layers rather than all in the output layer (plain-all) (ablation)DISCO · KITTI-3D 2D (All)PCK@0.1 (%): 80.6 -> 85.0Improved
supervise 3D shape concepts in intermediate layers rather than all in the output layer (plain-all) (ablation)DISCO · KITTI-3D 3D (Full)3D PCK@0.1 (%): 92.9 -> 95.3Improved
add more concepts in the correct order: DISCO-3D-2D to DISCO-vis-3D-2D to DISCO (ablation)DISCO · KITTI-3D 2D (All)PCK@0.1 (%): 80.7 -> 83.4 -> 85.0Improved
use the proposed supervision order instead of reversed order (ablation)DISCO · KITTI-3D 2D (All)PCK@0.1 (%): 25.6 -> 85.0Improved
use the proposed order instead of 3D-vis alternative order (ablation)DISCO · KITTI-3D 2D (All)PCK@0.1 (%): 84.5 -> 85.0Improved
remove local spatial pooling and use global average pooling instead of VGG-style (ablation)DISCO · KITTI-3D 2D (All)PCK@0.1 (%): 69.0 -> 85.0Improved
add simulated truncation and multi-car occlusion to training data (ablation)DISCO · KITTI-3D 2D (Truncation)PCK@0.1 (%): 53.6 -> 78.5Improved
add simulated truncation and multi-car occlusion to training data (ablation)DISCO · KITTI-3D 2D (Full)PCK@0.1 (%): 91.8 -> 93.1Improved
use RCNN detection boxes instead of ground-truth boxesDISCO · KITTI-3D 2D (All)PCK@0.1 (%): 85.0 -> 88.3Improved
compare synthetic-only DISCO to real-trained VpsKps on small imagesDISCO · PASCAL VOC car (Small Image)PCK@0.1 (%): 67.4 -> 74.3Improved

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