Multimodal feature generation framework for semantic image classification
Proceedings of the 2nd ACM International Conference on Multimedia Retrieval
Spherical soft assignment: improving image representation in content-based image retrieval
PCM'12 Proceedings of the 13th Pacific-Rim conference on Advances in Multimedia Information Processing
Comparison of mid-level feature coding approaches and pooling strategies in visual concept detection
Computer Vision and Image Understanding
Local hypersphere coding based on edges between visual words
ACCV'12 Proceedings of the 11th Asian conference on Computer Vision - Volume Part I
A comparative study of encoding, pooling and normalization methods for action recognition
ACCV'12 Proceedings of the 11th Asian conference on Computer Vision - Volume Part III
Exploring dense trajectory feature and encoding methods for human interaction recognition
Proceedings of the Fifth International Conference on Internet Multimedia Computing and Service
A local descriptor based on Laplacian pyramid coding for action recognition
Pattern Recognition Letters
Background subtraction using hybrid feature coding in the bag-of-features framework
Pattern Recognition Letters
Bilevel visual words coding for image classification
IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence
Towards large-scale geometry indexing by feature selection
Computer Vision and Image Understanding
| Hi-index | 0.00 |
In object recognition, soft-assignment coding enjoys computational efficiency and conceptual simplicity. However, its classification performance is inferior to the newly developed sparse or local coding schemes. It would be highly desirable if its classification performance could become comparable to the state-of-the-art, leading to a coding scheme which perfectly combines computational efficiency and classification performance. To achieve this, we revisit soft-assignment coding from two key aspects: classification performance and probabilistic interpretation. For the first aspect, we argue that the inferiority of soft-assignment coding is due to its neglect of the underlying manifold structure of local features. To remedy this, we propose a simple modification to localize the soft-assignment coding, which surprisingly achieves comparable or even better performance than existing sparse or local coding schemes while maintaining its computational advantage. For the second aspect, based on our probabilistic interpretation of the soft-assignment coding, we give a probabilistic explanation to the magic max-pooling operation, which has successfully been used by sparse or local coding schemes but still poorly understood. This probability explanation motivates us to develop a new mix-order max-pooling operation which further improves the classification performance of the proposed coding scheme. As experimentally demonstrated, the localized soft-assignment coding achieves the state-of-the-art classification performance with the highest computational efficiency among the existing coding schemes.