Machine Perception

Machine perception focuses on enabling computers to "see" and interpret the world like humans, primarily aiming to improve the accuracy and robustness of artificial vision systems. Current research emphasizes enhancing perception through integrating human behavioral data (e.g., eye-tracking, brainwaves) to guide machine learning models, developing efficient compression techniques tailored for machine vision tasks, and mitigating vulnerabilities to adversarial attacks and biases. These advancements are crucial for improving autonomous systems (robotics, self-driving cars), augmented reality applications, and addressing societal biases embedded in AI models, ultimately impacting various fields from manufacturing to art curation.