Ashish Patel 🇮🇳’s Post

𝗗𝗮𝘆-𝟰𝟭𝟯 𝗖𝗼𝗺𝗽𝘂𝘁𝗲𝗿 𝗩𝗶𝘀𝗶𝗼𝗻 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 Mirror-Yolo: An attention-based instance segmentation and detection model for mirrors by University of Liverpool, UK Follow me for a similar post: Ashish Patel ------------------------------------------------------------------- 𝗜𝗻𝘁𝗲𝗿𝗲𝘀𝘁𝗶𝗻𝗴 𝗙𝗮𝗰𝘁𝘀 : 🔸This paper is published arxiv2022. 👉 the Mirror-YOLO based on the YOLOv4 isproposed to address the problem of mirror detection, it comprises a structurally optimised attention mechanism module, the hypercolumn-stairstep module, and a bounding polygonsmodule to achieve instance segmentation and high accuracy in mirror detection. The paper also proposes a novel mirrorimage dataset with a low similarity rate and high robustness. Experiments have demonstrated the effectiveness of Mirror-YOLO compared to other state-of-the-art object detection net- works in mirror detection tasks. ------------------------------------------------------------------- 𝗜𝗠𝗣𝗢𝗥𝗧𝗔𝗡𝗖𝗘 ✔️ Mirrors can degrade the performance of computer vision models, however to accurately detect mirrors in images remains challenging. ✔️ YOLOv4 achieves phenomenal results both in object detection accuracy and speed, nevertheless the model often fails in detecting mirrors. ✔️ In this paper, a novel mirror detection method `Mirror-YOLO' is proposed, which mainly targets on mirror detection. ✔️ Based on YOLOv4, the proposed model embeds an attention mechanism for better feature acquisition, and a hypercolumn-stairstep approach for feature map fusion. ✔️ Mirror-YOLO can also produce accurate bounding polygons for instance segmentation. ✔️ The effectiveness of our proposed model is demonstrated by our experiments, compared to the existing mirror detection methods, the proposed Mirror-YOLO achieves better performance in detection accuracy on the mirror image dataset. #computervision #artificialintelligence #data