Dewi Mutiara Sari

About

Dewi Mutiara Sari is a researcher at the forefront of industrial robotic vision, specializing in 3D object detection, pose estimation, and bin-picking automation. Her work addresses critical challenges in manufacturing, where robots must accurately recognize and manipulate objects from cluttered, piled bins. Sari’s most cited paper, “Density-based Clustering for 3D Stacked Pipe Object Recognition using Directly-given Point Cloud Data on Convolutional Neural Network” (2022, 6 citations), introduces a novel recognition pipeline that integrates density-based clustering with CNNs to identify stacked 3D pipes directly from point cloud data—a key step often overlooked in bin-picking systems. She further advances pose estimation accuracy in “Optimization Estimating 3D Object Pose Using Levenberg-Marquardt Method” (2019, 2 citations), proposing a cost-effective solution using mono-vision cameras. Her 2022 work on “3D Object Detection Based on Point Cloud Data” (2 citations) continues this theme, focusing on robust detection of industrial pipes. With a total of 10 citations across her top papers, Sari’s contributions are vital for enabling efficient, low-cost robotic grasping in real-world manufacturing, making her a rising voice in applied computer vision and robotics.

Research Focus

Key Achievements

Top Papers

1. 1
   Density-based Clustering for 3D Stacked Pipe Object Recognition using Directly-given Point Cloud Data on Convolutional Neural Network

   6 citations · 2022
2. 2
   Optimization Estimating 3D Object Pose Using Levenberg-Marquardt Method

   2 citations · 2019
3. 3
   3D Object Detection Based on Point Cloud Data

   2 citations · 2022

Key Collaborators

Contact & Links