Scalable Graph Hashing with Feature Transformation Top-k Precision @MIRFLICKR-1M 08 0.7 0.8 0.7 0.6 0.6 05 0.5 SGH SGH +-T0 0.4 ITQ AGH 0.4 -AGH ◆DGH-可 ◆-DGH- 03 -DGH-R 0.3 -DGH-R PCAH PCAH ◆-LSH 0.2 ◆-LSH 02 200 400 600 800 1000 0 200 400 600 800 1000 #Returned Samples #Retumned Samples (a)64 bit (b)128 bit 口卡+得二4元互)Q0 Li (http://cs.nju.edu.cn/lvj) Learning to Hash LAMDA.CS.NJU 35 /43Scalable Graph Hashing with Feature Transformation Experiment Top-k Precision @MIRFLICKR-1M 0 200 400 600 800 1000 0.2 0.3 0.4 0.5 0.6 0.7 0.8 #Returned Samples Precision SGH ITQ AGH DGH−I DGH−R PCAH LSH (a) 64 bit 0 200 400 600 800 1000 0.2 0.3 0.4 0.5 0.6 0.7 0.8 #Returned Samples Precision SGH ITQ AGH DGH−I DGH−R PCAH LSH (b) 128 bit Li (http://cs.nju.edu.cn/lwj) Learning to Hash LAMDA, CS, NJU 35 / 43