Auston Sterling

University of North Carolina at Chapel Hill

Papers

Total Citations

H-Index

About

Auston Sterling is a researcher whose work lies at the intersection of computer vision, audio processing, and multimodal machine learning, with a particular focus on understanding everyday physical interactions. His most-cited paper, "Analyzing Liquid Pouring Sequences via Audio-Visual Neural Networks" (2019, 17 citations), introduces a novel approach to estimating the weight of poured liquids by leveraging audio data alongside visual inputs. This work addresses a key limitation in prior research, which often required predefined source weights or relied solely on visual cues. Sterling’s multimodal convolutional neural networks (CNNs) demonstrate that sound can effectively augment or even replace visual information for precise weight estimation, opening new avenues for applications in robotics, smart kitchens, and human activity recognition. By combining auditory and visual streams, his research advances the field of multimodal perception, showing how machines can better interpret complex, real-world actions. Sterling’s contributions highlight the power of integrating diverse sensory data to solve practical problems, making his work a valuable reference for students and researchers exploring sensor fusion and embodied AI.