AI Paper Atlas

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

Using Object Information for Spotting Text

Shitala Prasad, Adams Wai Kin Kong

ECCV 2018 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary object detection · 25 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 object-label pretraining and object features to the text detector (ablation)TO-CNN · NTU-UTOIF-score: 0.59 -> 0.66Improved
train Faster R-CNN on object and text labels together instead of text onlyFaster RCNN · NTU-UTOIF-score: 0.58 -> 0.37Hurt
add the third fine-tuning stage of the whole network (ablation)TO-CNN · NTU-UTOIF-score: 0.59 -> 0.66Improved
add the second training stage (Text VGG-16 on frozen object features) (ablation)TO-CNN · NTU-UTOIF-score: 0.42 -> 0.59Improved
use five aspect ratios instead of one with a single scale 8 (ablation)TO-CNN · NTU-UTOIF-score: 0.32 -> 0.58Improved
use three scales instead of one with a single ratio (ablation)TO-CNN · NTU-UTOIF-score: 0.32 -> 0.62Improved
combine three scales with five ratios instead of three scales and one ratio (ablation)TO-CNN · NTU-UTOIF-score: 0.62 -> 0.66Improved
improve anchors for Faster R-CNN pre-trained on NTU-UTOI objects (ablation)Faster R-CNN · NTU-UTOIF-score: 0.37 -> 0.57Improved
pretrain on NTU-UTOI objects instead of COCO objects before fine-tuning on text (ablation)Faster R-CNN (same anchors) · NTU-UTOIF-score: 0.24 -> 0.37Improved
additionally train on COCO-Text text labelsTO-CNN · COCO-TextF-score: 0.43 -> 0.45Improved
use object information (TO-CNN) versus prior best on recallTO-CNN · MSRA-TD500recall: 0.76 -> 0.90Improved

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