¥¼¥ß¥Ê¡¼¥ëȯɽ

Æü»þ: 12·î7Æü¡Ê²Ð¡Ë3¸Â (13:30-15:00)


²ñ¾ì: L1

»Ê²ñ¡§¶¶ËÜ¡¡·òÆó
ÃÝÇ÷¡¡À²Ç· ¥³¥ó¥Ô¥å¡¼¥Æ¥£¥ó¥°¡¦¥¢¡¼¥­¥Æ¥¯¥Á¥ã ÃæÅç¡¡¹¯É§
 
ÀµÌÚ¡¡¿Î ¥½¥Õ¥È¥¦¥§¥¢¹©³Ø ¾¾ËÜ¡¡·ò°ì
ȯɽÂêÌÜ¡§·²½°¥³¥ß¥å¥Ë¥±¡¼¥·¥ç¥ó¥¤¥ó¥¿¥Õ¥§¡¼¥¹¤Îɾ²Á
ȯɽ³µÍס§¶áǯ¡¤¥ª¥ó¥é¥¤¥ó¥³¥ß¥å¥Ë¥±¡¼¥·¥ç¥ó¤Î¿ÍͲ½¤Ëȼ¤¤¡¤¥ª¥ó¥é¥¤¥ó¾å¤Ç¶¦Ä̤δؿ´¤ò»ý¤Ä̵¿ô¤Î¿Í¡¹¤¬Æ±»þ¤Ë¥³¥ß¥å¥Ë¥±¡¼¥·¥ç¥ó¤ò¹Ô¤¦µ¡²ñ¤¬Áý²Ã¤·¤Æ¤¤¤ë¡¥¤·¤«¤·¤Ê¤¬¤é¡¤¥¤¥ó¥¿¡¼¥Í¥Ã¥È¤ÏƱ¤¸°Õ¸«¤ä´¶ÁÛ¤ò»ý¤Ä¿Í¡¹¤ò·ë¤ÓÉÕ¤±¤ëÆÃħ¤ò»ý¤Ä¤¿¤á¡¤¤·¤Ð¤·¤ÐÊиþŪ¤Ê°Õ¸«¤ä´¶ÁÛ¤¬»Ù»ý¤µ¤ì¤ë¤È¤¤¤¦ÌäÂ꤬¤¢¤ë¡¥Àè¹Ô¸¦µæ¤Ç¤Ï¡¤¥ª¥ó¥é¥¤¥ó¤Î·²½°¤ò¸½¼Â¼Ò²ñ¤Î·²½°¤ËÅö¤Æ¤Ï¤á¤ÆÄêµÁ¤·¡¤·²½°¥³¥ß¥å¥Ë¥±¡¼¥·¥ç¥ó¤ò»Ù±ç¤¹¤ë¤¿¤á¤Î¥¤¥ó¥¿¥Õ¥§¡¼¥¹¤ò»îºî¤·¡¤¤³¤ÎÌäÂê¤ò²ò·è¤·¤è¤¦¤È»î¤ß¤¿¡¥Ëܸ¦µæ¤Ç¤Ï¡¤Äó°Æ¥¤¥ó¥¿¥Õ¥§¡¼¥¹¤ÎÍ­ÍÑÀ­¤òɾ²Á¤¹¤ë¤¿¤á¤Ë¸½ºß¼çή¤Î¥Æ¥­¥¹¥È¥Ù¡¼¥¹¤Î¥¤¥ó¥¿¥Õ¥§¡¼¥¹¤È¤ÎÈæ³Ó¼Â¸³¤ò¹Ô¤¤¡¤Äó°Æ¥¤¥ó¥¿¥Õ¥§¡¼¥¹¤¬·²½°¥³¥ß¥å¥Ë¥±¡¼¥·¥ç¥ó¤ËµÚ¤Ü¤¹±Æ¶Á¤òÌÀ¤é¤«¤Ë¤¹¤ë¡¥ËÜȯɽ¤Ç¤Ï¡¤¼Â¸³¤Î³µÍפȺ£¸å¤ÎͽÄê¤Ë¤Ä¤¤¤ÆÊó¹ð¤¹¤ë¡¥
 
¼ÅÄ¡¡³¨Íý ¥³¥ó¥Ô¥å¡¼¥¿Àß·×³Ø Æ£¸¶¡¡½¨Íº
ȯɽÂêÌÜ¡§ÏÀʸ¾Ò²ð"Efficient Partial Scan Cell Gating for Low-Power Scan-Based Testing"
ȯɽ³µÍס§²óÏ©¤Î½Ð²Ù¥Æ¥¹¥È¤Ï¡¤À½ÉʤÎÉʼÁ¤òÊݾڤ¹¤ë¾å¤Ç½ÅÍפǤ¢¤ë¡¥½Ð²Ù¥Æ¥¹¥È¤òÍưפˤ¹¤ë¼êË¡¤Î°ì¤Ä¤Ë¥¹¥­¥ã¥óÀ߷פ¬¤¢¤ë¡¥¥¹¥­¥ã¥óÀ߷פ˴ð¤Å¤¯½Ð²Ù¥Æ¥¹¥È¤Ç¤Ï¡¤Ä̾ﵯ¤³¤ê¤¨¤Ê¤¤¾õÂÖÁ«°Ü¤¬µ¯¤­¤ë¤Ê¤É¡¤Ä̾ïÆ°ºî»þ¤è¤ê¤â¿¤¯¤ÎÅÅÎϤò¾ÃÈñ¤¹¤ë¡¥¤½¤Î·ë²Ì¡¤²óÏ©¤¬Â»½ý¤·¤¿¤ê¡¤À¸»º¥³¥¹¥È¤¬Áý²Ã¤¹¤ë¤Ê¤É¿·¤¿¤ÊÌäÂê¤ò°ú¤­µ¯¤³¤¹¡¥ËÜÏÀʸ¤Ç¤Ï¡¤²óÏ©Í×ÁǤòÁÞÆþ¤¹¤ë¤³¤È¤Ç¿®¹æÀþ¤ò£°Ëô¤Ï£±¤Ë¸ÇÄꤹ¤ë¥²¡¼¥Æ¥£¥ó¥°µ»½Ñ¤òÍѤ¤¤ë¤³¤È¤Ë¤è¤ê¡¤¥¹¥¤¥Ã¥Á¥ó¥°¤òÍÞÀ©¤·¡¤Æ°ÅªÅÅÎϤκ︺¤ò¼Â¸½¤·¤Æ¤¤¤ë¡¥ËÜȯɽ¤Ç¤Ï¡¤ÏÀʸ¾Ò²ð¤Ë²Ã¤¨¡¤¼«¿È¤Î¸¦µæ¤È¤Î´ØÏ¢¤Ë¤Ä¤¤¤Æ½Ò¤Ù¤ë¡¥
 
¿¹±Ê¡¡Íβ𠥳¥ó¥Ô¥å¡¼¥¿Àß·×³Ø Æ£¸¶¡¡½¨Íº
ȯɽÂêÌÜ¡§ÏÀʸ¾Ò²ð"Operating System Scheduling for Efficient Online Self-Test in Robust Systems"
ȯɽ³µÍס§¶áǯ¡¢¥Þ¥ë¥Á¥³¥¢¥×¥í¥»¥Ã¥µ¤¬ÉáµÚ¤·¤Æ¤¤¤ë¡£¿®ÍêÀ­¤Î¹â¤¤¥Þ¥ë¥Á¥³¥¢¥×¥í¥»¥Ã¥µ¤Ë¤Ï¡¢Ä̾ïÆ°ºî¤ò°Ý»ý¤·¤Æ¹Ô¤¦¥ª¥ó¥é¥¤¥ó¼«¸Ê¥Æ¥¹¥È¤¬É¬ÍפǤ¢¤ë¡£ ¤·¤«¤·¡¢¥³¥¢¤Î¥Æ¥¹¥È¤Ç¤Ï¤½¤Î¥³¥¢¤¬»ý¤Ä¥¿¥¹¥¯¤ò¤¹¤Ù¤Æ¥¹¥È¡¼¥ë¤·¤Æ¤Þ¤¦¡£¤½¤³¤Ç¡¢¤³¤ÎÏÀʸ¤Ç¤Ï¡¢¥ª¥ó¥é¥¤¥ó¼«¸Ê¥Æ¥¹¥È¤Ë¤ª¤¤¤Æ¡¢¥³¥¢¤¬»ý¤Ä¥¿¥¹¥¯¤Î°ÜÆ°¤ò¹Í¤¨¡¢ À­Ç½Îô²½¤òËɤ°ÊýË¡¤¬Äó°Æ¤µ¤ì¤Æ¤¤¤ë¡£ËÜȯɽ¤Ç¤Ï¡¢ÏÀʸ¤ÎÄó°Æ¼êË¡¤ò¾Ò²ð¤·¡¢»ä¤Î¸¦µæ¤È¤Î´Ø·¸¤Ë¤Ä¤¤¤Æ½Ò¤Ù¤ë¡£
 
µÈ²¬¡¡½ÓÊå ¥½¥Õ¥È¥¦¥§¥¢¹©³Ø ¾¾ËÜ¡¡·ò°ì
ȯɽÂêÌÜ¡§³«È¯·Ð¸³¤Ë¤è¤ëÆñ°×Å٤ΰۤʤ륽¡¼¥¹¥³¡¼¥É¤ÈÆɲò»þ´Ö¤Î±Æ¶ÁʬÀÏ
ȯɽ³µÍס§¶áǯ¡¤¥½¥Õ¥È¥¦¥§¥¢¤Î³«È¯¥µ¥¤¥¯¥ë¤Îû´ü²½¤Ëȼ¤¤¡¤´û¸¤Î¥½¥Õ¥È¥¦¥§¥¢¤ËÊÑ ¹¹¤ò²Ã¤¨¤Æ³«È¯¤ò¹Ô¤¦Êݼ鳫ȯ¤¬¹­¤¯ºÎÍѤµ¤ì¤Æ¤¤¤ë¡¥Êݼ鳫ȯ¤Ç¤Ï¡¤²þ¤Éôʬ ¡Ê¥Ñ¥Ã¥Á¡Ë¤òŬÍѤ¹¤ëÁ°¤Ë¡¤Â¾¤Î³«È¯¼Ô¤Ë¥Ñ¥Ã¥Á¤ò¥ì¥Ó¥å¡¼¤·¤Æ¤â¤é¤¦¤¬¡¤¥ì¥Ó ¥å¡¼ÂоÝʪ¤Ë¾Ü¤·¤¤ÆÃÄê¤Î³«È¯¼Ô¤Ë¥ì¥Ó¥å¡¼¤¬½¸Ã椹¤ë·¹¸þ¤Ë¤¢¤ë¡¥ ¡¡Ëܸ¦µæ¤ÎÌÜŪ¤Ï¡¤¥ì¥Ó¥å¡¼¤Î¥¿¥¹¥¯¤Î½¸Ãæ¡ÊÉԶѹաˤò²ò¾Ã¤¹¤ë¤¿¤á¤Ë¡¤¡Ö¥ì ¥Ó¥å¡¼Âоݡפȡ֥ì¥Ó¥å¡¼¤ò¹Ô¤¦³«È¯¼Ô¡×¤ÎÁȹç¤ï¤»¤ò·×²è¤¹¤ëºÝ¤Ë¹Íθ¤¹¤Ù¤­ Í×°ø¤òÌÀ¤é¤«¤Ë¤¹¤ë¤³¤È¤Ë¤¢¤ë¡¥¸¦µæ¤Î¥¢¥×¥í¡¼¥Á¤È¤·¤Æ¡¤³«È¯¤Î¼Â̳·Ð¸³¼Ô¤ò È︳¼Ô¤È¤·¤ÆÆñ°×Å٤ΰۤʤ뿼ïÎà¤Î¥Ñ¥Ã¥Á¤òÍѤ¤¤Æ¼Â¸³¤ò¹Ô¤¤¡¤Æɲò»þ´Ö¤ÈÀµ ÅúΨ¤òʬÀϤ¹¤ë¡¥
 
Devisetti Venkatarama Naveen ¥³¥ó¥Ô¥å¡¼¥Æ¥£¥ó¥°¡¦¥¢¡¼¥­¥Æ¥¯¥Á¥ã ÃæÅç¡¡¹¯É§
ȯɽÂêÌÜ¡§LAPP: An Energy-Efficient, High-Performance Accelerator.
ȯɽ³µÍס§ Many-core architectures take advantages of parallel programming methods while their performance gaining is at a cost of proportional power increase, which necessitates the need of per core power efficiency optimization. To solve this inconvenience, we propose an energy efficient, high-performance accelerator, named Linear Array Pipeline Processor (LAPP). LAPP improves performance by mapping instructions onto execution units of array, and are pipeline executed with data stream. LAPP can reduce its power via properly gating units during a long period. We developed LAPP simulator based on power reduction model and calculated the parameters of power. Compared to a Many-core of the same area of LAPP, the results indicated that the performance of LAPP is doubled, and the power consumption is reduced to 20%. LAPP achieved 10 times the power efficiency of Many cores.
 

²ñ¾ì: L2

»Ê²ñ¡§³ÀÆâ ÀµÇ¯
úóÈø¡¡¾©ÂÀ ¥¤¥ó¥¿¡¼¥Í¥Ã¥È¡¦¥¢¡¼¥­¥Æ¥¯¥Á¥ã º½¸¶¡¡½¨¼ù
ȯɽÂêÌÜ¡§¥­¥ã¥ó¥Ñ¥¹¥Í¥Ã¥È¥ï¡¼¥¯¤Ë¤ª¤±¤ë¥Î¡¼¥É´Æ»ë¥·¥¹¥Æ¥à
ȯɽ³µÍס§ ¥Í¥Ã¥È¥ï¡¼¥¯¤Ç¥¤¥ó¥·¥Ç¥ó¥È¤¬È¯À¸¤·¤¿ºÝ¤Ë¡¤¤½¤Î¥Û¥¹¥È¤Î½êºß¤òÌÀ¤é¤«¤Ë¤¹¤ë¤³¤È¤Ï½ÅÍפǤ¢¤ë¡¥¥¤¥ó¥·¥Ç¥ó¥È¤òȯÀ¸¤µ¤»¤¿¥Û¥¹¥È¤Î¾ì½ê¤ò¸Â¤é¤ì¤¿¾ðÊó¤ò¸µ¤ËÌÀ¤é¤«¤Ë¤·¤Ê¤±¤ì¤Ð¤Ê¤é¤Ê¤¤¡¥¤Þ¤¿¡¤¥¤¥ó¥·¥Ç¥ó¥ÈȯÀ¸¸å¤ËÊó¹ð¤ò¼õ¤±¤¿¾ì¹ç¤Ë¤Ï¡¤²áµî¤ËÁ̤äƥΡ¼¥É¾ðÊó¤ò¸¡º÷¤¹¤ëɬÍפ¬¤¢¤ë¡¥¤½¤³¤Ç¡¤¥­¥ã¥ó¥Ñ¥¹¥Í¥Ã¥È¥ï¡¼¥¯¤ËÀܳ¤µ¤ì¤¿¥Î¡¼¥É¤Î¾ðÊó¤ò¥Í¥Ã¥È¥ï¡¼¥¯¥¹¥¤¥Ã¥Á¤«¤é¼ý½¸¤·¡¤É¬Íפʻþ¤Ë¤½¤Î¾ðÊó¤ò¸¡º÷¤Ç¤­¤ë¥Î¡¼¥É´Æ»ë¥·¥¹¥Æ¥à¤ÎÀ߷פȹ½ÃÛ¤ò¹Ô¤Ã¤¿¡¥ËÜ¥¼¥ß¥Ê¡¼¥ëȯɽ¤Ç¤Ï¤½¤Î¥·¥¹¥Æ¥à¤Ë¤Ä¤¤¤ÆÀâÌÀ¤¹¤ë¡¥
 
À¾³À¡¡·Ë ¥¤¥ó¥¿¡¼¥Í¥Ã¥È¡¦¥¢¡¼¥­¥Æ¥¯¥Á¥ã º½¸¶¡¡½¨¼ù
ȯɽÂêÌÜ¡§·×»»µ¡¼¼¤Ë¤ª¤±¤ë¾ÃÈñÅÅÎϴƻ륷¥¹¥Æ¥à
ȯɽ³µÍס§·×»»µ¡´ï¤Î¾ÃÈñÅÅÎϤÏǯ¡¹Áý²Ã·¹¸þ¤Ë¤¢¤ë¤È¤¤¤ï¤ì¤Æ¤¤¤ë¡¥´Ä¶­ÌäÂê¤Î¿¼¹ï²½¤Ë¤È¤â¤Ê¤¤¡¤·×»»µ¡±¿ÍѤˤ«¤«¤ëÅÅÎϤòºï¸º¤·¤¿¤¤¤È¤¤¤¦Í׵᤬¤¢¤ë¡¥¾ÊÅÅÎϲ½¤Î¤¿¤á¤Ë¤Ï¡¤¼ÂºÝ¤Ë¤É¤ì¤À¤±¤ÎÅÅÎϤò¾ÃÈñ¤·¤Æ¤¤¤ë¤Î¤«ÇÄ°®¤¹¤ë¤³¤È¤¬½ÅÍפǤ¢¤ë¡¥¤½¤³¤Ç¡¤Ëܸ¦µæ¤Ç¤Ï·×»»µ¡¼¼¤Î¾ÃÈñÅÅÎϤβĻ벽¤ò¹Ô¤¦¡¥¤Þ¤¿²Ä»ë²½¤·¤¿¥Ç¡¼¥¿¤ò¸µ¤Ë¸úΨŪ¤Ê¾ÊÅÅÎϲ½¼êË¡¤Ë¤Ä¤¤¤Æ¹Í»¡¤¹¤ë¡¥
 
ÉÛ°ú¡¡Í¤Íè ¥¤¥ó¥¿¡¼¥Í¥Ã¥È¡¦¥¢¡¼¥­¥Æ¥¯¥Á¥ã º½¸¶¡¡½¨¼ù
ȯɽÂêÌÜ¡§¥¯¥é¥¦¥É¥³¥ó¥Ô¥å¡¼¥Æ¥£¥ó¥°¤ò»Ù¤¨¤ë ¥Í¥Ã¥È¥ï¡¼¥¯¶¦Í­¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à¤ÎÀ­Ç½É¾²Á
ȯɽ³µÍס§¥¤¥ó¥¿¡¼¥Í¥Ã¥È¤ò³èÍѤ·¤¿¥¯¥é¥¦¥É¥³¥ó¥Ô¥å¡¼¥Æ¥£¥ó¥°¤¬ÃíÌܤµ¤ì¤Æ¤ª¤ê¡¤¤½¤ÎÃæ¤Ç¤â¥¯¥é¥¦¥É¥³¥ó¥Ô¥å¡¼¥Æ¥£¥ó¥°¤Ë¤ª¤±¤ë¥¹¥È¥ì¡¼¥¸¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à¥µ¡¼¥Ó¥¹¤Î¼ûÍפ¬¹â¤Þ¤Ã¤Æ¤¤¤ë¡¥Ëܸ¦µæ¤Ç¤Ï¡¤¥¯¥é¥¦¥É¥³¥ó¥Ô¥å¡¼¥Æ¥£¥ó¥°¸þ¤±¤Î¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à¤ò¹½ÃÛ¤·¡¤À­Ç½É¾²Á¤ò¹Ô¤¦¡¥¤Þ¤¿¡¤¥¢¥¯¥»¥¹¥Ñ¥¿¡¼¥ó¤ò¹Íθ¤·¤¿¥Ù¥ó¥Á¥Þ¡¼¥¯¥½¥Õ¥È¤Ê¤É¤Î¼ÂÁõ¤ò¹Ô¤¤¡¤¹½ÃÛ¤·¤¿¥·¥¹¥Æ¥à¾å¤ÇÀ­Ç½É¾²Á¤ò¹Ô¤¦¤³¤È¤Ç¡¤Gfarm ¤ä HDFS ¤Ê¤É¤Î¥Í¥Ã¥È¥ï¡¼¥¯¶¦Í­¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à¥¢¡¼¥­¥Æ¥¯¥Á¥ã¤Ë¤Ä¤¤¤Æ¹Í»¡¤ò¹Ô¤¦¡¥
 
Æ£ËÜ¡¡¶³Ê¿ ¥½¥Õ¥È¥¦¥§¥¢´ðÁÃ³Ø °ËÆ£¡¡¼Â
ȯɽÂêÌÜ¡§°ÜÆ°¥ª¥Ö¥¸¥§¥¯¥È¤ÎÄÉÀפò¹Ô¤¦ÌµÀþ¥Þ¥ë¥Á¥á¥Ç¥£¥¢¥»¥ó¥µ¥Í¥Ã¥È¥ï¡¼¥¯¤ÎĹ¼÷Ì¿²½¤òÌܻؤ¹¥ë¡¼¥Æ¥£¥ó¥°¼êË¡
ȯɽ³µÍס§Ëܸ¦µæ¤Ç¤Ï¡¤°ÜÆ°¥ª¥Ö¥¸¥§¥¯¥È¤ò¸¡ÃΡ¦»£±Æ¤·¡¤¤½¤Î±ÇÁü¤ò¡¤¥Þ¥ë¥Á¥Û¥Ã¥×̵ÀþÄÌ¿®¤ò¹Ô¤Ã¤Æ¥æ¡¼¥¶¤¬½ê»ý¤¹¤ë·ÈÂÓüËö¤Ë¥ê¥¢¥ë¥¿¥¤¥à¤ËÆϤ±¤ë̵Àþ¥Þ¥ë¥Á¥á¥Ç¥£¥¢¥»¥ó¥µ¥Í¥Ã¥È¥ï¡¼¥¯¡ÊWMSN¡Ë¤òÄó°Æ¤¹¤ë¡¥¥ª¥Ö¥¸¥§¥¯¥ÈÄÉÀפò¹Ô¤¦WMSN¤Ç¤Ï¡¤ÌµÀþ¥»¥ó¥µ¥Î¡¼¥É¤Î¸Â¤é¤ì¤¿»ñ¸»¤òÍѤ¤¤Æ¡¤Æ°²è¤ÎÉʼÁ¤äÇÛÁ÷¤Î¥ê¥¢¥ë¥¿¥¤¥àÀ­¤È¤¤¤Ã¤¿QoS¤ò¼Â¸½¤Ç¤­¤ë¤«¤¬²ÝÂê¤È¤Ê¤ë¡¥¤Þ¤¿¡¤ÄÉÀפ¹¤Ù¤­¥ª¥Ö¥¸¥§¥¯¥È¤Î¿ô¤Ë±þ¤¸¤ÆÃæ·Ñ¥È¥é¥Õ¥£¥Ã¥¯¤¬ÁýÂ礷¡¤ÅÅÇȤδ³¾Ä¤Ë¤è¤ëíÕíÔ¤¬µ¯¤³¤ê¤ä¤¹¤¯¤Ê¤ê¡¤Æ°²èÉʼÁ¤ÎÄã²¼¤ò¾·¤¯¡¥¤µ¤é¤Ë¡¤¾ï¤ËºÇû¥Ñ¥¹¤ÇÆ°²è¤òÃæ·Ñ¤¹¤ë¤È°ìÉô¤Î¥Î¡¼¥É¤Î¥Ð¥Ã¥Æ¥ê¤¬Â¾¤è¤êÁ᤯¸Ï³é¤·¤Æ¤·¤Þ¤¤¡¤WMSNÁ´ÂΤβÔƯ»þ´Ö¤¬Ã»¤¯¤Ê¤ë¤È¤¤¤Ã¤¿ÌäÂ꤬ȯÀ¸¤¹¤ë¡¥Ëܸ¦µæ¤Ç¤Ï¡¤¥Í¥Ã¥È¥ï¡¼¥¯Á´ÂΤβÔƯ»þ´ÖºÇÂç²½¤ÈQoS³ÎÊݤòƱ»þ¤Ë¼Â¸½¤¹¤ë¥ë¡¼¥Æ¥£¥ó¥°¼êË¡¤òÄó°Æ¤¹¤ë¡¥¤½¤Î¤¿¤á¤Ë¡¤¥ª¥Ö¥¸¥§¥¯¥È¤È¥æ¡¼¥¶¤È¤Îµ÷Î¥¤Ë±þ¤¸¤ÆµöÍƲÄǽ¤ÊÇÛÁ÷ÃÙ±ä¤òÄê¤á¡¤¤½¤ÎÈÏ°ÏÆâ¤ÇÅÅÃÓ»ÄÎ̤¬¾¯¤Ê¤¤¥Î¡¼¥É¤ò±ª²ó¤¹¤ë¤è¤¦¤ÊÇÛÁ÷·ÐÏ©¤òÁªÂò¤¹¤ë¤³¤È¤Ç¡¤¥»¥ó¥µ¥Î¡¼¥É´Ö¤Î¾ÃÈñÅÅÎÏÎ̤ζÑÅù²½¤È¥È¥é¥Õ¥£¥Ã¥¯½¸Ãæ¤Ë¤è¤ëíÕíԤβóÈò¤ò¿Þ¤ë¡¥
 
¾¾üâ¡¡Áï»Ë ¥¤¥ó¥¿¡¼¥Í¥Ã¥È¡¦¥¢¡¼¥­¥Æ¥¯¥Á¥ã º½¸¶¡¡½¨¼ù
ȯɽÂêÌÜ¡§Â絬ÌÏ¥¦¥§¥Ö¥µ¥¤¥È¤Î±¿ÍѤȤ½¤Î¥¢¥¯¥»¥¹¥Ñ¥¿¡¼¥ó²òÀÏ
ȯɽ³µÍס§Âè92²óÁ´¹ñ¹â¹»ÌîµåÁª¼ê¸¢Âç²ñ¤Î´Ö, »î¹ç·Ð²áÅù¤òÇÛ¿®¤¹¤ëÂ絬ÌÏ¥¦¥§¥Ö¥µ¥¤¥È¹½ÃÛ¡¦±¿ÍѤò¹Ô¤Ã¤¿. ¤Þ¤¿±¿ÍѤòÄ̤¸¤Æ¼ý½¸¤·¤¿¥¢¥¯¥»¥¹¥í¥°¤ò´ð¤Ë¥¢¥¯¥»¥¹¥Ñ¥¿¡¼¥ó²òÀϤò¹Ô¤¤, ¤½¤ÎÆÃħ¤òÄ´ºº¤·¤¿. ËÜ¥¼¥ß¥Ê¡¼¥ëȯɽ¤Ç¤Ï¤½¤ÎÂ絬ÌÏÇÛ¿®¥¦¥§¥Ö¥µ¥¤¥È¤Î¥·¥¹¥Æ¥à¹½À®¤ª¤è¤Ó, ¥¢¥¯¥»¥¹¥Ñ¥¿¡¼¥ó¤ÎÆÃħ¤Ë¤Ä¤¤¤Æ¹Í»¡¤¹¤ë.
 

²ñ¾ì: L3

»Ê²ñ¡§ÃÝÇ·Æâ ¹â»Ö
ÅÄÃæ¡¡ÂÙ»Ë Áü¾ðÊó½èÍý³Ø À鸶¡¡Ô¢¹¨
ȯɽÂêÌÜ¡§ÏÀʸ¾Ò²ð"Modeling 3D Human Poses from Uncalibrated Monocular Images"
ȯɽ³µÍס§¶áǯ¡¢Ã±´ã¥«¥á¥é²èÁü¤«¤é¤Î3¼¡¸µ¹½Â¤ºÆ¹½À®¤Ë´Ø¤¹¤ë¸¦µæ¤¬¥³¥ó¥Ô¥å¡¼¥¿¥Ó¥¸¥ç¥ó¤ÎʬÌî¤ÇÀ¹¤ó¤Ë¹Ô¤ï¤ì¤Æ¤¤¤ë¡£¤·¤«¤·½¾Íè¤Î¸¦µæ¤Ç¤Ï¡¢ºÆ¹½À®·ë²Ì¤ÎÛ£Ë椵¤ä¿¤¯¤Î»öÁ°Ã챤òɬÍפȤ¹¤ë¤Ê¤É¤ÎÌäÂ꤬¤¢¤Ã¤¿¡£¤³¤ÎÏÀʸ¤Ç¤Ï¥­¥ã¥ê¥Ö¥ì¡¼¥·¥ç¥ó¤µ¤ì¤Æ¤¤¤Ê¤¤2¼¡¸µÃ±´ã²èÁü·²¤Î¾¯¤Ê¤¤ÂбþÅÀ¤«¤é3¼¡¸µ¤Î¿ÍÂλÑÀª¤È¥«¥á¥é¥Ñ¥é¥á¡¼¥¿¤òƱ»þ¤Ë¹â¤¤ÀºÅ٤Ǻƹ½À®¤¹¤ë¸úΨŪ¤Ê¥¢¥ë¥´¥ê¥º¥à¤ò¾Ò²ð¤·¤Æ¤¤¤ë¡£ËÜȯɽ¤Ç¤Ï¤³¤ÎÏÀʸ¤ò¾Ò²ð¤¹¤ë¤È¤È¤â¤Ë¡¢¼«Ê¬¤Î¸¦µæ¤È¤Î´ØÏ¢¤ò½Ò¤Ù¤ë¡£
 
ÅÄÃæ¡¡¹¨µ¨ ¸À¸ì²Ê³Ø ¼¯Ìî¡¡À¶¹¨
ȯɽÂêÌÜ¡§¼«Êľɻù»Ù±ç¤Î¤¿¤á¤Î¾Ð¤¤Ê¬ÀÏ
ȯɽ³µÍס§¼«Êľɻù¤ÏÂÐÏÃÁê¼ê¤Îµ¤»ý¤Á¤äµ÷Î¥´¶¤ÎÍý²ò¤òÆä˶ì¼ê¤È¤·¤Æ¤¤¤ë¡£²æ¡¹¤Ï¼«Êľɻù¤Ë¤½¤ì¤é¤òÅÁ¤¨¤ë¼êÃʤȤ·¤Æ¾Ð¤¤¤ËÃíÌܤ¹¤ë¡£¾Ð¤¤¤Î¼ïÎà¤ò¼«Æ°¼±Ê̤·ÅÁ㤹¤ë¤³¤È¤òÌÜŪ¤È¤·¡¢Â絬Ìϼ«Á³ÂÐÏÃ¥³¡¼¥Ñ¥¹¤Î¾Ð¤¤¤Î²»¶ÁÆÃħÎ̤ˤĤ¤¤ÆʬÀϤ¹¤ë¡£ËÜȯɽ¤Ç¤Ï¡¢PCA¡¢Classification Tree¤òÍѤ¤¤¿Ê¬ÀϤȡ¢SVM¼«Æ°¼±Ê̤ˤĤ¤¤Æ¤Î¼Â¸³·ë²Ì¤òÊó¹ð¤¹¤ë¡£
 
ÅÄÃæ¡¡²Â¼ù »ë³Ð¾ðÊó¥á¥Ç¥£¥¢ ²£Ìð¡¡Ä¾ÏÂ
ȯɽÂêÌÜ¡§[ÏÀʸ¾Ò²ð]Accurate, Dense, and Robust Multi-View Stereopsis
ȯɽ³µÍס§ËÜÏÀʸ¤Ç¤Ï¡¢ÆþÎϲèÁü¤«¤é¹âÀºÅÙ¡¢¹âÌ©ÅÙ¡¢¥í¥Ð¥¹¥ÈÀ­¤ò¤â¤Ä»°¼¡¸µ¾ðÊó¤ÎÉü¸µË¡¤òÄó°Æ¤¹¤ë¡£Äó°Æ¼êË¡¤Ï¡¢»°¼¡¸µÉü¸µ¤ÎÃæ¤Î°ì¼êË¡¤Ç¤¢¤ë¿´ã¥¹¥Æ¥ì¥ª(Multi-View Stereo)¤òÍѤ¤¤¿¡¢¥Ñ¥Ã¥Á¥Ù¡¼¥¹(patch-based)¤Î¼êË¡¤Ç¤¢¤ë¡£½èÍý¤Î³µÍפȤ·¤Æ¤Ï¡¢¥Ñ¥Ã¥Á¤Î¥Þ¥Ã¥Á¥ó¥°¡¢³ÈÄ¥¡¢¥Õ¥£¥ë¥¿¥ê¥ó¥°¤Î½èÍý¤ò·«¤êÊÖ¤·¹Ô¤Ã¤¿¸å¡¢¥Ý¥ê¥´¥ó¥á¥Ã¥·¥å¤È¤·¤Æ½ÐÎϤ¹¤ë¡£Äó°Æ¼êË¡¤Ï¡¢ÄºÅÀ¤ÎË¡ÀþÊý¸þ¤ò¿äÄꤹ¤ë¤³¤È¤¬ÆÃħ¤Ç¤¢¤ë¡£ÄêÎÌŪɾ²Á¤Ç¤Ï¡¢6¹àÌܤΤ¦¤Á4¹àÌܤǾ¼êË¡¤è¤êÍ¥¤ì¤¿ÀºÅÙ¤ò¼¨¤·¤Æ¤ª¤ê¡¢ÍÍ¡¹¤Ê¥·¡¼¥ó¤ÎÉü¸µ¤ËÂФ·¤ÆÍ­¸ú¤Ç¤¢¤ë¤³¤È¤¬¼¨¤µ¤ì¤¿¡£»°¼¡¸µÉü¸µ¤Î°ì¼êË¡¤È¤·¤Æ¡¢ËÜÏÀʸ¤ÎÄó°Æ¼êË¡¤Î¾Ò²ð¤ò¹Ô¤¦¤È¤È¤â¤Ëº£¸å¤Î¸¦µæ¤È¤Î´ØÏ¢¤Ë¤Ä¤¤¤Æ½Ò¤Ù¤ë¡£
 
¾ïµÈ¡¡¹â¹° ¼«Á³¸À¸ì½èÍý³Ø ¾¾ËÜ¡¡Íµ¼£
ȯɽÂêÌÜ¡§Wikipedia¤òÍѤ¤¤¿¥ª¥ó¥È¥í¥¸¡¼¹½Ãۤ˴ؤ¹¤ë¸¦µæ
ȯɽ³µÍס§¥ª¥ó¥È¥í¥¸¡¼¤È¸Æ¤Ð¤ì¤ë¡Ö³µÇ°´Ö¤Î´Ø·¸¤Ê¤É¤ò½¸¤á¤¿µ¡³£²ÄÆɤʼ­½ñ¡×¤¬Â¸ºß¤¹¤ë¡£¤³¤Î¥ª¥ó¥È¥í¥¸¡¼¤òÍøÍѤ¹¤ë¤³¤È¤Ç¡¤Web¤Î½ÀÆð¤Ê¸¡º÷¤¬¤Ç¤­¤ë¤è¤¦¤Ë¤Ê¤ë¤È¹Í¤¨¤é¤ì¤ë¡£¤·¤«¤·¡¤¥ª¥ó¥È¥í¥¸¡¼¤ò¿Í¼ê¤Ç¹½ÃÛ¡¦³ÈÄ¥¤¹¤ë¤Î¤Ë¤ÏÈó¾ï¤Ë¥³¥¹¥È¡¦»þ´Ö¤¬¤«¤«¤ê¡¤¸½¾õ¡¤¸Çͭ̾»ì¤ä¿·¤·¤¤¸ì¤Ë¤Ä¤¤¤Æ¤Ï¤¢¤Þ¤êÀ°È÷¤µ¤ì¤Æ¤¤¤Ê¤¤¾õ¶·¤Ë¤¢¤ë¡£¤½¤³¤Ç¡¤¼«Á³¸À¸ì½èÍý¤òÍѤ¤¤Æ¼«Æ°Åª¤Ë¥ª¥ó¥È¥í¥¸¡¼¤ò¹½ÃÛ¡¦³ÈÄ¥¤¹¤ë¼êË¡¤¬¿ô¿¤¯Äó°Æ¤µ¤ì¤Æ¤¤¤ë¡£ËÜȯɽ¤Ç¤Ï¡¤¼«Á³¸À¸ì½èÍý¤òÍѤ¤¤¿¥ª¥ó¥È¥í¥¸¡¼¹½ÃÛ¡¦³ÈÄ¥¤Ë´Ø¤¹¤ëÀè¹Ô¸¦µæ¤ò¾Ò²ð¤·¡¤º£¸å¤Î¼«¤é¤Î¸¦µæÊý¿Ë¤Ë¤Ä¤¤¤Æ½Ò¤Ù¤ë¡£
 
Ë­Àî¡¡¹°¼ù ¸À¸ì²Ê³Ø ¼¯Ìî¡¡À¶¹¨
ȯɽÂêÌÜ¡§ÇÈ·ÁÀܳ·¿²»À¼¹çÀ®Ë¡¤ò²þÎɤ·¤¿¹âÉʼÁ²»À¼¹çÀ®¼êË¡¤ÎÄó°Æ
ȯɽ³µÍס§¹âÉʼÁ²½¤¬µá¤á¤é¤ì¤ë²»À¼¹çÀ®Ë¡¤Î1¤Ä¤Ç¤¢¤ëÇÈ·ÁÀܳ²»À¼¹çÀ®Ë¡¤Ï¡¤¹çÀ®²»À¼¤ËÀܳÏĤߤ¬À¸¤¸¤ë¤È¤¤¤¦ÌäÂ꤬¤¢¤ë¡¥ËÜÄó°Æ¼êË¡¤Ç¤Ïʸ¾ÏÁ´ÂΤËÅϤäÆñ°Ì¤ÎÁ°¸å´Ä¶­¤ª¤è¤Ó±¤Î§ÅªÆÃħ¤ò¹Íθ¤¹¤ë¤³¤È¤Ç¡¤¹çÀ®²»À¼¤ÎÉʼÁ¸þ¾å¤òÌܻؤ¹¡¥ËÜȯɽ¤Ç¤Ï¡¤Äó°Æ¼êË¡¤Î¼Â¸½¤ÎÁ°Ãʳ¬¤È¤·¤Æ»Ò²»¤ÎÁ°¸åÊì²»´Ä¶­¤ËÃíÌܤ·¡¤¹çÀ®²»À¼¤òºîÀ®¤·¤Æɾ²Á¤ò¹Ô¤Ã¤¿¡¥