Abstract:Semantic correspondence remains a challenging task for establishing correspondences between a pair of images with the same category or similar scenes due to the large intra-class appearance. In this paper, we introduce a novel problem called 'Small Object Semantic Correspondence (SOSC).' This problem is challenging due to the close proximity of keypoints associated with small objects, which results in the fusion of these respective features. It is difficult to identify the corresponding key points of the fused features, and it is also difficult to be recognized. To address this challenge, we propose the Keypoint Bounding box-centered Cropping (KBC) method, which aims to increase the spatial separation between keypoints of small objects, thereby facilitating independent learning of these keypoints. The KBC method is seamlessly integrated into our proposed inference pipeline and can be easily incorporated into other methodologies, resulting in significant performance enhancements. Additionally, we introduce a novel framework, named KBCNet, which serves as our baseline model. KBCNet comprises a Cross-Scale Feature Alignment (CSFA) module and an efficient 4D convolutional decoder. The CSFA module is designed to align multi-scale features, enriching keypoint representations by integrating fine-grained features and deep semantic features. Meanwhile, the 4D convolutional decoder, based on efficient 4D convolution, ensures efficiency and rapid convergence. To empirically validate the effectiveness of our proposed methodology, extensive experiments are conducted on three widely used benchmarks: PF-PASCAL, PF-WILLOW, and SPair-71k. Our KBC method demonstrates a substantial performance improvement of 7.5\% on the SPair-71K dataset, providing compelling evidence of its efficacy.

Exploiting Feature Correspondence Constraints for Image Recognition

Image Classification Method by Combining Multi-features and Sparse Coding

Hierarchical Clustering Matching for Features with Repetitive Patterns in Visual Odometry

Sparse Codebook Model of Local Structures for Retrieval of Focal Liver Lesions Using Multiphase Medical Images

A More Robust Feature Correspondence for more Accurate Image Recognition

A Branch-and-Bound Approach to Correspondence and Grouping Problems

Scalable local feature matching without visual codebook training

Constrained Branch-and-Bound Algorithm for Image Registration

Object Recognition via Adaptive Multi-level Feature Integration

Accurate Image Search with Multi-Scale Contextual Evidences

Feature-matching method based on keypoint response constraint using binary encoding of phase congruency

Independently Keypoint Learning for Small Object Semantic Correspondence

Unifying Discriminative Visual Codebook Generation with Classifier Training for Object Category Recognition

Supervised Sparse Patch Coding Towards Misalignment-Robust Face Recognition

Neural Best-Buddies: Sparse Cross-Domain Correspondence

A data-driven study of image feature extraction and fusion

Discriminative Codeword Selection for Image Representation.

Permutation-invariant Feature Restructuring for Correlation-aware Image Set-based Recognition

Image Recognition Based on Compressive Imaging and Optimal Feature Selection

Adaptively feature matching via joint transformational-spatial clustering

Inter-Image Communication for Weakly Supervised Localization