AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Functionality Discovery and Prediction of Physical Objects

Lei Ji, Botian Shi, Xianglin Guo, Xilin Chen

AAAI 2020 · Detection and segmentation › Open-set and few-shot detection · approach: Hand-object and articulated interaction · 3 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 consolidation of ConceptNet, pattern and model-based tuplesHead-Action extraction · Head-Action datasetF1: 0.580 -> 0.610Improved
use consolidation instead of ConceptNet onlyHead-Action extraction · Head-Action datasetrecall: 0.152 -> 0.624Improved
use consolidation instead of ConceptNet onlyHead-Action extraction · Head-Action datasetprecision: 0.862 -> 0.597Hurt
use model-based instead of pattern-based extractionHead-Action extraction · Head-Action datasetprecision: 0.632 -> 0.694Improved
use model-based instead of pattern-based extractionHead-Action extraction · Head-Action datasetrecall: 0.536 -> 0.255Hurt
directly classify functionality from image instead of object classification + PRAobject-functionality prediction · Head(image)-Action-Tail testF1: 0.486 -> 0.756Improved
use combined PRA+Dismult inference after object classification instead of functionality classificationobject-functionality prediction · Head(image)-Action-Tail testF1: 0.756 -> 0.7789Improved
combine functionality classification with combined inference scoresobject-functionality prediction · Head(image)-Action-Tail testF1: 0.7789 -> 0.810Improved
use Dismult instead of PRA after object classificationobject-functionality prediction · Head(image)-Action-Tail testrecall: 0.388 -> 0.633Improved
add object classification to functionality classification (joint model) (ablation)visual module · functionality classificationF1: 0.734 -> 0.731No clear effect
train Dismult on consolidated graph instead of ConceptNet (ablation)Dismult module · object-functionality predictionF1: 0.539 -> 0.651Improved
use consolidated graph instead of ConceptNet for PRA (ablation)PRA module · random sampleF1: 0.134 -> 0.552Improved

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