Compositional Convolutional Neural Networks: A Robust and Interpretable Model for Object Recognition Under Occlusion

By Adam Kortylewski, Qing Liu, Angtian Wang, Yihong Sun, Alan L. Yuille. 2020

90 citations recorded by the source corpus.

Computer vision systems in real-world applications need to be robust to partial occlusion while also being explainable. In this work, we show that black-box deep convolutional neural networks (DCNNs) have only limited robustness to partial occlusion. We overcome these limitations by unifying DCNNs with part-based models into Compositional Convolutional Neural Networks (CompositionalNets)—an interpretable deep architecture with innate robustness to partial occlusion. Specifically, we propose to replace the fully connected classification head of DCNNs with a differentiable compositional model that can be trained end-to-end. The structure of the compositional model enables CompositionalNets to decompose images into objects and context, as well as to further decompose object representations in terms of individual parts and the objects’ pose. The generative nature of our compositional model enables it to localize occluders and to recognize objects based on their non-occluded parts. We conduct extensive experiments in terms of image classification and object detection on images of artificially occluded objects from the PASCAL3D+ and ImageNet dataset, and real images of partially occluded vehicles from the MS-COCO dataset. Our experiments show that CompositionalNets made from several popular DCNN backbones (VGG-16, ResNet50, ResNext) improve by a large margin over their non-compositional counterparts at classifying and detecting partially occluded objects. Furthermore, they can localize occluders accurately despite being trained with class-level supervision only. Finally, we demonstrate that CompositionalNets provide human interpretable predictions as their individual components can be understood as detecting parts and estimating an objects’ viewpoint.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Philip H. S. Torr
2. Alan L. Yuille
3. Xinggang Wang
4. Song Bai
5. Xiang Bai
6. Serge J. Belongie
7. Bernt Schiele
8. Trevor Darrell

In Computer Vision and Pattern Recognition

1. Xinggang Wang
2. Xiang Bai
3. Alan L. Yuille
4. Luc Van Gool
5. Chunhua Shen
6. Philip H. S. Torr
7. Ming-Hsuan Yang
8. Dacheng Tao

In Safety, Robustness & Interpretability

1. Song Bai
2. Philip H. S. Torr
3. Serge J. Belongie
4. Alan L. Yuille

Citations in the corpus

• Occluded Video Instance Segmentation: A Benchmark (2021, methodology): “Different from the full-DNN paradigm described above, Some methods[40,41,42] integrate compositional models and deep convolutional neural networks into a unified model which is more robust to partial occlusions.”
• Occluded Video Instance Segmentation: Dataset and ICCV 2021 Challenge (2021, background): “[21] integrates compositional models and deep convolutional neural networks into a unified model which is more robust to partial occlusions.”