AI Paper Atlas

Research map / Video / Video-language understanding

Pivot Correlational Neural Network for Multimodal Video Categorization

Sunghun Kang, Junyeong Kim, Hyunsoo Choi, Sungjin Kim, Chang D. Yoo

ECCV 2018 · Video › Video-language understanding · approach: Temporal video grounding · 10 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
replace GRU with contextual GRU (cGRU) for the pivot hidden state (ablation)Pivot CorrNN baseline · FCVID (C3D + VGGish)mAP (%): 66.86 -> 67.57Improved
add pivot correlation maximization on hidden state (ablation)Pivot CorrNN with cGRU · FCVID (C3D + VGGish)mAP (%): 67.57 -> 67.68Improved
add pivot correlation maximization on prediction (ablation)Pivot CorrNN with cGRU · FCVID (C3D + VGGish)mAP (%): 67.57 -> 68.02Improved
maximize correlation on both hidden state and prediction (ablation)Pivot CorrNN with cGRU · FCVID (C3D + VGGish)mAP (%): 68.02 -> 68.45Improved
add adaptive aggregation module (ablation)Pivot CorrNN with cGRU and pivot correlation · FCVID (C3D + VGGish)mAP (%): 68.45 -> 69.54Improved
add all three proposed modules to the GRU baseline (ablation)GRU baseline · FCVID (C3D + VGGish)mAP (%): 66.86 -> 69.54Improved
add appearance features (CNN, SIFT) (ablation)Pivot CorrNN · FCVID testmAP (%): 69.54 -> 75.33Improved
add motion features IDT-HOF, IDT-HOG (ablation)Pivot CorrNN · FCVID testmAP (%): 75.33 -> 76.58Improved
add motion features IDT-MBH, IDT-Traj (ablation)Pivot CorrNN · FCVID testmAP (%): 76.58 -> 77.23Improved
add MFCC audio features (ablation)Pivot CorrNN · FCVID testmAP (%): 77.23 -> 77.60Improved
use Pivot CorrNN instead of best prior method rDNNFCVID classifier · FCVID testmAP (%): 76.0 -> 77.6Improved
use Pivot CorrNN without cGRU instead of logistic regression on concatenated featuresvideo-level features · YouTube-8M (video-level features)GAP (%): 76.79 -> 77.40Improved

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