# Hierarchical Salient Object Detection for Assisted Grasping

## TL;DR

This paper presents a hierarchical saliency detection method derived from scene segmentation to improve object detection and assist robotic grasping, demonstrating effective scene understanding and manipulation capabilities.

## Contribution

It introduces a novel hierarchical saliency function based on segmentation, enabling better detection of salient objects and regions for robotic grasping tasks.

## Key findings

- Effective detection of salient objects in complex scenes
- Hierarchical saliency provides scale-specific region identification
- Successful implementation in a pick-and-place robotic system

## Abstract

Visual scene decomposition into semantic entities is one of the major challenges when creating a reliable object grasping system. Recently, we introduced a bottom-up hierarchical clustering approach which is able to segment objects and parts in a scene. In this paper, we introduce a transform from such a segmentation into a corresponding, hierarchical saliency function. In comprehensive experiments we demonstrate its ability to detect salient objects in a scene. Furthermore, this hierarchical saliency defines a most salient corresponding region (scale) for every point in an image. Based on this, an easy-to-use pick and place manipulation system was developed and tested exemplarily.