dual view hashing

Dual-view Hashing MOHAMMAD RASTEGARI JONGHYUN CHOI SHOBEIR - PowerPoint PPT Presentation

Predictable Dual-view Hashing MOHAMMAD RASTEGARI JONGHYUN CHOI SHOBEIR FAKHRAEI* HAL DAUM III LARRY S. DAVIS University of Maryland, College Park, MD, USA Integrate information from different modalities Text - Image Human Riding


  1. Predictable Dual-view Hashing MOHAMMAD RASTEGARI JONGHYUN CHOI SHOBEIR FAKHRAEI* HAL DAUMÉ III LARRY S. DAVIS University of Maryland, College Park, MD, USA

  2. Integrate information from different modalities  Text - Image Human Riding Horse  Sound - Text ICML 2013 | Atlanta | Predictable Dual-view Hashing

  3. Dual-View Hashing Textual Space Visual Space Human riding a horse Person laughing at restaurant Cow standing in a farm Hamming Space 010111010 101010101 110010101 ICML 2013 | Atlanta | Predictable Dual-view Hashing

  4. Binary code assignments with splits 0 1 1 0 001 1 0 101 ICML 2013 | Atlanta | Predictable Dual-view Hashing

  5. Predictability 14 01?101 12 10 8 6 4 2 0 -2 -20 -15 -10 -5 0 5 10 15 20 25 ICML 2013 | Atlanta | Predictable Dual-view Hashing

  6. Predictability 11 0 110 14 10 0 110 01 ? 101 12 01 0 110 01 0 011 10 8 6 4 2 0 -2 -20 -15 -10 -5 0 5 10 15 20 25 ICML 2013 | Atlanta | Predictable Dual-view Hashing

  7. Predictability 14 CCA 12 10 8 6 4 2 0 -2 -20 -15 -10 -5 0 5 10 15 20 25 ICML 2013 | Atlanta | Predictable Dual-view Hashing

  8. Predictability 14 PDH CCA 12 10 8 6 margin 4 2 0 -2 -20 -15 -10 -5 0 5 10 15 20 25 ICML 2013 | Atlanta | Predictable Dual-view Hashing

  9. Optimization Solved by a block coordinate descent algorithm ICML 2013 | Atlanta | Predictable Dual-view Hashing

  10. Predictable dual-view hashing Modality 1 Modality 2 1101 0101 0010 1110 label 1010 1010 0101 0101 1110 1001 ICML 2013 | Atlanta | Predictable Dual-view Hashing

  11. Predictable dual-view hashing Modality 1 Modality 2 1101 0101 0010 1110 label 1010 1010 0101 0101 1110 1001 ICML 2013 | Atlanta | Predictable Dual-view Hashing

  12. Image  image (with 32 bits) ICML 2013 | Atlanta | Predictable Dual-view Hashing

  13. Text  Image (with 32 bits) ICML 2013 | Atlanta | Predictable Dual-view Hashing

  14. Image Retrieval Pascal Sentence [Farhadi et al] Sun Attribute [Peterson et al] ICML 2013 | Atlanta | Predictable Dual-view Hashing

  15. Thank you Predictable Dual-view Hashing Mohammad Rastegari, Jonghyun Choi, Shobeir Fakhraei*, Hal Daumé III and Larry S. Davis University of Maryland, College Park, MD, USA

Recommend


More recommend