Search papers, labs, and topics across Lattice.
This study introduces a unified tokenization framework that processes heterogeneous 3D modalities for pain recognition, integrating behavioral and brain-activity data into a single pipeline. By preserving the spatial, temporal, and time-frequency structures of diverse inputs, the framework eliminates the need for separate architectures and handcrafted biases. Extensive evaluations on the AI4Pain benchmark demonstrate that this approach achieves state-of-the-art performance while ensuring computational efficiency for real-time applications on both GPU and CPU platforms.
A unified framework for pain recognition achieves state-of-the-art results by seamlessly integrating diverse 3D data modalities without the need for separate processing architectures.
Pain is a complex and pervasive phenomenon affecting a large percentage of the population, and accurate assessment is essential for effective clinical management and intervention. Computational pain recognition systems enable continuous monitoring, support clinical decision-making, and help mitigate pain-related distress and functional decline. This study introduces a unified tokenization framework for heterogeneous 3D modalities in pain recognition that provides a single processing pipeline across behavioral and brain-activity 3D data, without requiring separate architectures for each modality or handcrafted inductive biases. The framework preserves spatial, temporal, and time--frequency structure while mapping diverse inputs into a shared token space. Extensive experiments show that the proposed approach effectively processes facial videos and fNIRS data in both raw-signal and spectrogram-based representations. On the AI4Pain benchmark dataset, the proposed framework achieves state-of-the-art performance while maintaining high computational efficiency and enabling real-time assessment on both GPU and CPU hardware.
