Domain Adaptation for Visual Understanding

"Domain Adaptation for Visual Understanding" (red. Richa Sing, Mayank Vatsa, Vishal M. Patel, Nalini Ratha) omhandler den seneste udvikling af maskinlæringsalgortimer inden for visuel forståelse og genkendelse.

 

I "Domain Adaptation for Visual Understanding" præsenterer en række eksperter deres perspektiver på visuel forståelse, objekt- og ansigtsgenkendelse samt handlings- og adfærdsgenkendelse inden for maskinlæringsalgoritmer.                     

 

"Domain Adaptation for Visual Understanding" består af følgende ni kapitler:

1.Domain Adaptation for Visual Understanding, 2.M-ADDA: Unsupervised Domain Adaptation with Deep Metric Learning, 3. XGAN: Unsupervised Image-to-Image Translation for Many-to-Many Mappings, 4. Improving Transferability of Deep Neural Networks, 5. Cross-Modality Video Segment Retrieval with Ensemble Learning, 6.On Minimum Discrepancy Estimation for Deep Domain Adaptation, 7. Multi-modal Conditional Feature Enhancement for Facial Action Unit Recognition, 8. Intuition Learning, 9. Alleviating Tracking Model Degradation Using Interpolation-Based Progressive Updating

 

Richa Sing, Mayank Vatsa, Vishal M. Patel og Nalini Ratha har redigeret bogen.