AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Query-Guided Regression Network With Context Policy for Phrase Grounding

Kan Chen, Rama Kovvuri, Ram Nevatia

ICCV 2017 · Vision-language and multimodal › Human and embodied modeling · approach: Multimodal retrieval-augmented modeling · 120 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 retrieval GroundeR with query-guided regression (QRN) on RPN proposals (ablation)RPN proposals, VGG-det · Flickr30K Entitiesaccuracy (%): 39.13 -> 53.48Improved
replace retrieval GroundeR with QRN on Selective Search proposals (ablation)SS proposals, VGG-det · Flickr30K Entitiesaccuracy (%): 47.81 -> 55.99Improved
use jointly trained proposal generation network (PGN) instead of Selective Search (ablation)QRN · Flickr30K Entitiesaccuracy (%): 55.99 -> 60.21Improved
add context policy network (CPN) with reinforcement-learning rewards (ablation)PGN+QRN · Flickr30K Entitiesaccuracy (%): 60.21 -> 65.14Improved
add QRN regression and CPN on EdgeBoxes/PGN proposalsQRC Net · Referit Gameaccuracy (%): 26.93 -> 44.07Improved
use PGN instead of EdgeBoxes for QRN (ablation)QRN · Referit Gameaccuracy (%): 32.21 -> 43.57Improved
add CPN (ablation)PGN+QRN · Referit Gameaccuracy (%): 43.57 -> 44.07Improved
use PGN proposals instead of RPN proposals (ablation)proposal generation · Flickr30K EntitiesUBP (%): 71.25 -> 89.61Improved
increase regression loss weight lambda from 1.0 to 10.0 (ablation)QRC Net · Flickr30K Entitiesaccuracy (%): 65.14 -> 63.27Hurt
increase multimodal dimension m from 512 to 1024 (ablation)QRC Net · Flickr30K Entitiesaccuracy (%): 65.14 -> 62.52Hurt
increase CPN reward beta from 0.2 to 0.8 (ablation)QRC Net · Flickr30K Entitiesaccuracy (%): 65.14 -> 62.77Hurt
vary lambda from 0.5 to 4.0 (ablation)QRC Net · Flickr30K Entitiesaccuracy (%): 64.15 / 65.14 / 64.40 / 64.29No clear effect

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