A Unified Interactive Model Evaluation for Classification, Object Detection, and Instance Segmentation in Computer Vision

Changjian Chen, Yukai Guo, Fengyuan Tian, Shilong Liu, Weikai Yang, Zhaowei Wang, Jing Wu, Hang Su, Hanspeter Pfister, Shixia Liu

DOI: 10.1109/TVCG.2023.3326588

Room: 103

2023-10-24T22:48:00ZGMT-0600Change your timezone on the schedule page
2023-10-24T22:48:00Z
Exemplar figure, described by caption below
Uni-Evaluator is an open-source visual analysis tool to support a unified interactive model evaluation for classification, object detection, and instance segmentation in computer vision. The tool consists of (a) the filtering panel; (b) the matrix-based visualization that provides an overview of model performance; (c) the table visualization that helps identify problematic data subsets; and (d) the grid visualization that displays the samples of interest. These modules work together to facilitate the model evaluation from a global overview to individual samples.
Fast forward
Full Video
Keywords

Model evaluation, computer vision, classification, object detection, instance segmentation

Abstract

Existing model evaluation tools mainly focus on evaluating classification models, leaving a gap in evaluating more complex models, such as object detection. In this paper, we develop an open-source visual analysis tool, Uni-Evaluator, to support a unified model evaluation for classification, object detection, and instance segmentation in computer vision. The key idea behind our method is to formulate both discrete and continuous predictions in different tasks as unified probability distributions. Based on these distributions, we develop 1) a matrix-based visualization to provide an overview of model performance; 2) a table visualization to identify the problematic data subsets where the model performs poorly; 3) a grid visualization to display the samples of interest. These visualizations work together to facilitate the model evaluation from a global overview to individual samples. Two case studies demonstrate the effectiveness of Uni-Evaluator in evaluating model performance and making informed improvements.