Optimized clothes segmentation to boost gender classification in unconstrained scenarios
This work addresses the challenge of gender classification in unconstrained environments, which is important for applications like surveillance or marketing, but it is incremental as it builds on existing segmentation and classification methods.
The paper tackled the problem of demographic information extraction in unconstrained scenarios by developing a clothes segmentation method using trixels and a modified GrabCut algorithm, which when combined with face detection, achieved near real-time performance and significantly improved gender classification accuracy on the ClothesDB dataset.
Several applications require demographic information of ordinary people in unconstrained scenarios. This is not a trivial task due to significant human appearance variations. In this work, we introduce trixels for clustering image regions, enumerating their advantages compared to superpixels. The classical GrabCut algorithm is later modified to segment trixels instead of pixels in an unsupervised context. Combining with face detection lead us to a clothes segmentation approach close to real time. The study uses the challenging Pascal VOC dataset for segmentation evaluation experiments. A final experiment analyzes the fusion of clothes features with state-of-the-art gender classifiers in ClothesDB, revealing a significant performance improvement in gender classification.