Memetically Optimized MCWLD for Matching Sketches With Digital Face Images
Citations
4 citations
4 citations
3 citations
Cites methods from "Memetically Optimized MCWLD for Mat..."
...In the second approach, the discriminative methods utilize feature descriptors such as the scale-invariant feature transform (SIFT) [18], Weber’s local descriptor (WLD) [5], and multi-scale local binary pattern (MLBP) [11]....
[...]
3 citations
3 citations
Cites background or methods from "Memetically Optimized MCWLD for Mat..."
...State of the art approaches can be classified on the basis of the technique used to tackle the previously mentioned modality-gap in two main categories: generative or discriminative [9]....
[...]
...Unfortunately it is rather difficult to find public face datasets feasible in this scenario, so we created a mixed database of 8221 images taking well-controlled photos from various sources: 188 images from CUHK [4], 123 from AR [4], 65 from IIITD semi forensic [9], 114 from CVL [47], 100 from PUT [48], 1194 from Feret [6][7], 6387 from FRGC [49] and 50 real mug shots collected from the web....
[...]
...Previously mentioned approaches [20], [9], [8] and [24] share the ability to deal with viewed sketches as well as forensic sketches (differences between them are exhaustively described in [8])....
[...]
...Among the mentioned state of the art approaches, only a few works report results on so large datasets ([9][8][28])....
[...]
References
46,906 citations
14,708 citations
Additional excerpts
...On the other hand, sparse descriptor such as Scale Invariant Feature Transform (SIFT ) [23] is based on interest point detection and computing the descriptor in the vicinity of detected interest points....
[...]
14,157 citations
5,563 citations