Dual-Layer Density Estimation for Multiple Object Instance Detection

Figure 1

Overview of the proposed framework: (a) offline phase for constructing the retrieval structure; (b)–(e) first layer of density estimation: (b) local feature detection, (c) feature matching and key point mapping, (d) first layer of density estimation, and (e) intermediate results; (f)–(i) second layer of density estimation: (f) feature template reconstruction, (g) false matching result elimination, and (h) clustering for candidate instances detection; (i) geometric verification.