Material Classification in the Wild: Do Synthesized Training Data Generalise Better than Real-World Training Data?
This addresses the challenge of data scarcity and generalization in material classification for computer vision applications, though it is incremental as it builds on existing CNN architectures.
The paper tackles the problem of material classification by questioning whether synthesized training data generalizes better than real-world data, finding that synthesized data improves mean average precision by 5% to 19% across three real-world databases.
We question the dominant role of real-world training images in the field of material classification by investigating whether synthesized data can generalise more effectively than real-world data. Experimental results on three challenging real-world material databases show that the best performing pre-trained convolutional neural network (CNN) architectures can achieve up to 91.03% mean average precision when classifying materials in cross-dataset scenarios. We demonstrate that synthesized data achieve an improvement on mean average precision when used as training data and in conjunction with pre-trained CNN architectures, which spans from ~ 5% to ~ 19% across three widely used material databases of real-world images.