Yen‐Yu Chang

Papers

Total Citations

H-Index

About

Yen‐Yu Chang is a leading researcher in multisensory object-centric learning, with a focus on bridging the gap between simulated and real-world robotic perception. Their major contributions center on creating high-fidelity, multimodal datasets that integrate visual, auditory, and tactile representations of everyday objects. Chang’s pioneering work on the ObjectFolder series has been instrumental in advancing sim-to-real transfer for embodied AI. The first ObjectFolder dataset (2021, 23 citations) introduced 100 virtualized objects with implicit multisensory representations, addressing the limitations of geometry-centric synthetic data. Building on this, ObjectFolder 2.0 (2022, 59 citations) significantly enhanced realism and scale, enabling more robust cross-modal learning and robotic interaction. These datasets have become foundational resources for researchers studying multisensory perception and object manipulation, with their impact reflected in growing citation counts. Chang’s work is notable for tackling the critical challenge of unrealistic object modeling in prior research, providing the community with standardized benchmarks that accelerate progress in multisensory AI and robotics.