AI Paper Atlas

Research map / Detection and segmentation / CNN features

Dynamic Multi-Scale Filters for Semantic Segmentation

Junjun He, Zhongying Deng, Yu Qiao

ICCV 2019 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 321 citations (OpenAlex)

Read the paper

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 context-aware dynamic filters instead of fixed regular-conv filters (ablation)DCM · PASCAL VOC 2012 val, kernel sizes {1}mIoU (%): 70.76 -> 77.75Improved
use context-aware dynamic filters instead of fixed regular-conv filters (ablation)DCM · PASCAL VOC 2012 val, kernel sizes {1,3,5,7}mIoU (%): 72.37 -> 78.21Improved
use context-aware filters instead of fixed depthwise filters (ablation)DCM · PASCAL VOC 2012 val, kernel sizes {1,3}mIoU (%): 71.12 -> 78.52Improved
use context-aware filters instead of dilated filters {6,12,24,36} (ablation)DCM · PASCAL VOC 2012 valmIoU (%): 77.60 -> 78.21Improved
use context-aware filters instead of dilated filters {1,2,3,4} (ablation)DCM · PASCAL VOC 2012 valmIoU (%): 73.37 -> 78.21Improved
add more DCM branches (kernel sets {1} to {1,3,5}) (ablation)DMNet · PASCAL VOC 2012 valmIoU (%): 77.75 -> 78.76Improved
add a fourth DCM branch (kernel 7) to {1,3,5} (ablation)DMNet · PASCAL VOC 2012 valmIoU (%): 78.76 -> 78.21Hurt
generate filters from Stage5 instead of Stage2 context (ablation)DMNet · PASCAL VOC 2012 valmIoU (%): 73.13 -> 78.21Improved
add DCMs to FCN (ResNet50) (ablation)FCN · PASCAL VOC 2012 valmIoU (%): 69.83 -> 78.21Improved
add DCMs to FCN (ResNet101) (ablation)FCN · PASCAL VOC 2012 valmIoU (%): 76.24 -> 80.82Improved
use ResNet101 instead of ResNet50DMNet · PASCAL VOC 2012 valmIoU (%): 78.21 -> 80.82Improved
add deep supervision (ablation)DMNet ResNet101 · PASCAL VOC 2012 valmIoU (%): 80.87 -> 81.08Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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