In April 2025, Iyim developed an image segmentation feature for the ECLAIR-Robotics/crackle repository, focusing on precise object segmentation within bounding boxes using the Segment Anything Model (SAM). Iyim implemented segment_img.py and integrated it with yolo_segment_node.py, enabling the extraction of object-specific points to enhance perception accuracy in robotics applications. The work was carried out in Python and Shell, leveraging computer vision and ROS skills to improve the perception pipeline’s robustness and reusability. By reducing manual labeling effort and preparing the foundation for future SAM-driven features, Iyim delivered a focused, well-integrated solution that advanced object manipulation readiness in the system.