A Systematic Evaluation of Object Detection Networks for Scientific Plots

TL;DR

This paper evaluates the effectiveness of existing object detection networks on scientific plots, identifies their limitations especially at high localization precision, and proposes novel modifications to improve accuracy and efficiency for reasoning tasks.

Contribution

The paper introduces a series of modifications including a Laplacian edge-based region proposal method and a custom loss function, significantly enhancing detection accuracy and speed for scientific plot analysis.

Findings

Achieved 93.44% mAP at 0.9 IOU with the final model.

Model inference time is 16 times faster than existing detectors.

Performance on text objects remains a challenge.

Abstract

Are existing object detection methods adequate for detecting text and visual elements in scientific plots which are arguably different than the objects found in natural images? To answer this question, we train and compare the accuracy of various SOTA object detection networks on the PlotQA dataset. At the standard IOU setting of 0.5, most networks perform well with mAP scores greater than 80% in detecting the relatively simple objects in plots. However, the performance drops drastically when evaluated at a stricter IOU of 0.9 with the best model giving a mAP of 35.70%. Note that such a stricter evaluation is essential when dealing with scientific plots where even minor localisation errors can lead to large errors in downstream numerical inferences. Given this poor performance, we propose minor modifications to existing models by combining ideas from different object detection networks. While this significantly improves the performance, there are still 2 main issues: (i) performance on text objects which are essential for reasoning is very poor, and (ii) inference time is unacceptably large considering the simplicity of plots. To solve this open problem, we make a series of contributions: (a) an efficient region proposal method based on Laplacian edge detectors, (b) a feature representation of region proposals that includes neighbouring information, (c) a linking component to join multiple region proposals for detecting longer textual objects, and (d) a custom loss function that combines a smooth L1-loss with an IOU-based loss. Combining these ideas, our final model is very accurate at extreme IOU values achieving a mAP of 93.44%@0.9 IOU. Simultaneously, our model is very efficient with an inference time 16x lesser than the current models, including one-stage detectors. With these contributions, we enable further exploration on the automated reasoning of plots.