[Legacy Report] Computer Vision Series Talk: ACCV Awarded Paper - Sparse Coding on Cascade Residuals

#Sparse #Coding #computer #vision #machine #learning

Abstract: This paper seeks to combine dictionary learning and hier- archical image representation in a principled way. To make dictionary atoms capturing additional information from extended receptive fields and attain improved descriptive capacity, we present a two-pass multi- resolution cascade framework for dictionary learning and sparse coding. The cascade allows collaborative reconstructions at different resolutions using the same dimensional dictionary atoms. Our jointly learned dic- tionary comprises atoms that adapt to the information available at the coarsest layer where the support of atoms reaches their maximum range and the residual images where the supplementary details progressively refine the reconstruction objective. The residual at a layer is computed by the difference between the aggregated reconstructions of the previous lay- ers and the downsampled original image at that layer. Our method gen- erates more flexible and accurate representations using much less num- ber of coefficients. Its computational efficiency stems from encoding at the coarsest resolution, which is minuscule, and encoding the residuals, which are relatively much sparse. Our extensive experiments on multiple datasets demonstrate that this new method is powerful in image cod- ing, denoising, inpainting and artifact removal tasks outperforming the state-of-the-art techniques.

• Add Event to Calendar
  iCal
  Google Calendar

• Canberra, Australian Capital Territory
• Australia

---

• Contact Event Host
• Co-sponsored by youshaodi@gmail.com