New Graph Regularized Sparse Coding Improving Automatic Image Annotation

Typical image classification pipeline for shallow architecture can be summarized by the following three main steps: i) a projection in high dimensional space of local features, ii) sparse constraints for the encoding scheme and iii) a pooling operation to obtain a global representation invariant to common transformation. Sparse Coding (SC) framework is one particular example of this general approach. The main problem raised by it is the local feature encoding which is done independently, loosing correlation of the input space. In this work we propose to simultaneously encode sparse codes to tackle this problem with Joint Sparse Coding (JSC) inspired by Graph regularized Sparse Coding (GSC). We experiment SC, GSC and JSC on UIUCsports and scenes15 database. We will show that results obtained, for UIUCsports, with SC (87.27± 1.33), JSC (84.17±1.57) and the State-of-the-Art (88.47±2.32 [23]) are tackled by a simple fusion (95.37± 1.29). Several assumptions will be advanced to explain this phenomenon which can’t be generalized.