AI Paper Atlas

Research map / Detection and segmentation / CNN features

Joint Semantic Instance Segmentation on Graphs with the Semantic Mutex Watershed

Steffen Wolf, Yuyan Li, Constantin Pape, Alberto Bailoni, Anna Kreshuk, Fred A. Hamprecht

ECCV 2020 · Detection and segmentation › CNN features · approach: Boundary-aware instance segmentation

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 joint partitioning and labeling (SMWS) instead of MWS followed by max semantic edge labelingMutex Watershed · 3D sponge EM volume (panoptic)PQ: 48.1 -> 51.6Improved
use SMWS instead of connected components of semantic predictionsCCsem · 3D sponge EM volume (panoptic)PQ: 43.4 -> 51.6Improved
use SMWS instead of connected components of short-range affinitiesCCaff · 3D sponge EM volume (panoptic)PQ: 24.3 -> 51.6Improved
use SMWS instead of the Asymmetric Multiway Cut (KLj*r) solver on largest feasible 2D sliceAMWC · Sponge 2D slicePQ: 52.0 -> 49.2Hurt
use SMWS instead of PFPN with Mask-RCNN affinities and Deeplab semanticsPFPN · Cityscapes val (panoptic)PQ: 58.1 -> 59.3Improved
use SMWS instead of SSAPSSAP · Cityscapes val (panoptic)PQ: 61.1 -> 59.3Hurt
use SMWS instead of AdaptISAdaptIS · Cityscapes val (panoptic)PQ: 62.0 -> 59.3Hurt
use SMWS instead of UPSNetUPSNet · Cityscapes val (panoptic)PQ: 59.3 -> 59.3No clear effect
combine all affinity sources (Mask-RCNN, GMIS, Deeplab) instead of the best pair (ablation)SMWS · Cityscapes val (panoptic)PQ: 59.3 -> 46.3Hurt
derive affinities from Mask-RCNN foreground probabilities rather than other sources (ablation)SMWS · Cityscapes val (panoptic)PQImproved
replace AMWC solver by SMWS (p to infinity) for runtimeAMWC · Sponge 2D slicesruntime: 5 orders of magnitude fasterImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 3 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