Advances in Acoustic and Photoacoustic Imaging

The recent advancements in the field of acoustic and photoacoustic imaging have shown significant progress in enhancing the capabilities of these technologies, particularly in challenging environments and clinical applications. Researchers are increasingly focusing on integrating acoustic signals with traditional imaging modalities to overcome limitations such as occlusion, poor lighting, and privacy concerns. This integration is leading to more robust and versatile imaging solutions, as evidenced by innovations in 3D human mesh reconstruction and contact localization. Additionally, there is a notable shift towards developing data-free and tuning-free frameworks for denoising and artifact removal, which are crucial for real-time applications and clinical utility. These approaches not only improve the signal-to-noise ratio and image quality but also maintain the integrity of spectral information, making them highly effective in dynamic and challenging imaging conditions. Notably, zero-shot self-supervised methods are emerging as a promising direction, offering efficient and practical solutions for artifact removal without the need for extensive training data or prior knowledge. These developments collectively push the boundaries of what is possible with acoustic and photoacoustic imaging, paving the way for more reliable and advanced imaging technologies in the future.

Noteworthy papers include:

  • SonicBoom, which introduces a learning-based approach for contact localization using an array of microphones, demonstrating robust performance in visually challenging environments.
  • SonicMesh, combining acoustic signals with RGB images for 3D human mesh reconstruction, achieving high accuracy in challenging conditions.
  • SPADE, a data-free enhancement framework for spectroscopic photoacoustic imaging, improving SNR and preserving spectral information.
  • Zero-Shot Artifact2Artifact, a self-supervised method for artifact removal in photoacoustic imaging, achieving state-of-the-art performance with minimal computational requirements.

Sources

SonicBoom: Contact Localization Using Array of Microphones

Sonicmesh: Enhancing 3D Human Mesh Reconstruction in Vision-Impaired Environments With Acoustic Signals

SPADE: Spectroscopic Photoacoustic Denoising using an Analytical and Data-free Enhancement Framework

Zero-Shot Artifact2Artifact: Self-incentive artifact removal for photoacoustic imaging without any data

Built with on top of