Humans need not label more humans: Occlusion Copy & Paste for Occluded Human Instance Segmentation

TL;DR

This paper introduces Occlusion Copy & Paste, a data-centric augmentation method that improves occluded human instance segmentation by synthetically creating occluded examples, achieving state-of-the-art results without additional data or complex model changes.

Contribution

The paper presents a simple, effective data augmentation technique for occluded human segmentation that enhances performance without extra data or model modifications.

Findings

Improves occluded human segmentation performance significantly.

Achieves state-of-the-art results on OCHuman dataset.

Works with any existing instance segmentation model.

Abstract

Modern object detection and instance segmentation networks stumble when picking out humans in crowded or highly occluded scenes. Yet, these are often scenarios where we require our detectors to work well. Many works have approached this problem with model-centric improvements. While they have been shown to work to some extent, these supervised methods still need sufficient relevant examples (i.e. occluded humans) during training for the improvements to be maximised. In our work, we propose a simple yet effective data-centric approach, Occlusion Copy & Paste, to introduce occluded examples to models during training - we tailor the general copy & paste augmentation approach to tackle the difficult problem of same-class occlusion. It improves instance segmentation performance on occluded scenarios for "free" just by leveraging on existing large-scale datasets, without additional data or manual labelling needed. In a principled study, we show whether various proposed add-ons to the copy & paste augmentation indeed contribute to better performance. Our Occlusion Copy & Paste augmentation is easily interoperable with any models: by simply applying it to a recent generic instance segmentation model without explicit model architectural design to tackle occlusion, we achieve state-of-the-art instance segmentation performance on the very challenging OCHuman dataset. Source code is available at https://github.com/levan92/occlusion-copy-paste.