A5064273339- Advanced Image Fusion Techniques
- Face and Expression Recognition
- Remote-Sensing Image Classification
- Face recognition and analysis
- Sparse and Compressive Sensing Techniques
- Advanced Image Processing Techniques
- Image and Signal Denoising Methods
- Image Processing Techniques and Applications
- Photoacoustic and Ultrasonic Imaging
Xidian University
2021-2024
Matching infrared (IR) facial probes against a gallery of visible light faces remains challenge, especially when combined with cross-distance due to deteriorated quality the IR data. In this paper, we study scenario where are acquired at short standoff, while long-range To address issue imbalance between heterogeneous imagery, propose compensate it by upgrading lower-quality faces. Specifically, is realized through cascaded face enhancement that combines an existing denoising algorithm...
Target detection and recognition of bistatic SAR image has been widely studied in recently. However, how to accurately detect recognize targets with low-resolution small size the is still a problem. Image super-resolution reconstruction technology can increase resolution expand target improve performance. Thus, we this paper bring deep learning onto topic propose novel model (named FSRCNN) for images. The proposed characterized by feature extractor different structures, feedback enhancement...
Face recognition has been well studied under visible light and infrared (IR) in both intra-spectral cross-spectral cases. However, how to fuse different bands for face recognition, i.e., hyperspectral is still an open research problem, which the advantages of richer information retention all-weather functionality over single-band recognition. Thus, this research, we revisit problem provide a deep learning-based approach. A new fusion model (named HyperFace) proposed address problem. The...
Face recognition has already been well studied under the visible light and infrared,in both intra-spectral cross-spectral cases. However, how to fuse different bands, i.e., hyperspectral face recognition, is still an open research problem, which advantages of richer information retaining all-weather functionality over single band recognition. Among very few works for traditional non-deep learning techniques are largely used. Thus, we in this paper bring deep into topic propose a new fusion...