Face recognition across pose is a problem of fundamental importance in computer vision. We propose to address this problem by using stereo matching to judge the similarity of two, 2D images of faces seen from different poses. Stereo matching allows for arbitrary, physically valid, continuous correspondences. We show that the stereo matching cost provides a very robust measure of similarity of faces that is insensitive to pose variations. To enable this, we show that, for conditions common in face recognition, the epipolar geometry of face images can be computed using either four or three feature points. We also provide a straightforward adaptation of a stereo matching algorithm to compute the similarity between faces. The proposed approach has been tested on the CMU PIE data set and demonstrates superior performance compared to existing methods in the presence of pose variation. It also shows robustness to lighting variation.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TPAMI.2009.123 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Reliable Disparity Estimation Using Multiocular Vision with Adjustable Baseline.
Sensors (Basel)
December 2024
Institute for Computer Research, University of Alicante, P.O. Box 99, 03080 Alicante, Spain.
Article Synopsis
- Accurate 3D information estimation from images is crucial for computer vision, and while binocular stereo vision is a common approach, it faces challenges with baseline distance affecting reliability.
- This research proposes a new method that progressively increases the baseline in multiocular vision, introducing a rectification technique that significantly reduces distortion errors in the images.
- The method enhances disparity estimation accuracy by 20% for multiocular images and demonstrates superior performance through extensive evaluations against existing methods.
Stereo-seq of the prefrontal cortex in aging and Alzheimer's disease.
Nat Commun
January 2025
Tulane Center for Biomedical Informatics and Genomics, Deming Department of Medicine, School of Medicine, Tulane University, New Orleans, LA, 70112, USA.
Aging increases the risk for Alzheimer's disease (AD), driving pathological changes like amyloid-β (Aβ) buildup, inflammation, and oxidative stress, especially in the prefrontal cortex (PFC). We present the first subcellular-resolution spatial transcriptome atlas of the human prefrontal cortex (PFC), generated with Stereo-seq from six male AD cases at varying neuropathological stages and six age-matched male controls. Our analyses revealed distinct transcriptional alterations across PFC layers, highlighted disruptions in laminar structure, and exposed AD-related shifts in layer-to-layer and cell-cell interactions.
View Article and Find Full Text PDFDetecting somatic variants in purified brain DNA obtained from surgically implanted depth electrodes in epilepsy.
Epilepsia
January 2025
Department of Clinical Neurosciences, University of Calgary, Calgary, Alberta, Canada.
Objective: Somatic variants causing epilepsy are challenging to detect, as they are only present in a subset of brain cells (e.g., mosaic), resulting in low variant allele frequencies.
View Article and Find Full Text PDFStereo-Differentiating Asymmetric Rh(I)-Catalyzed Pauson-Khand Reaction: A DFT-Informed Approach to Thapsigargin Stereoisomers.
J Am Chem Soc
January 2025
Department of Chemistry, University of Pittsburgh, Pittsburgh, Pennsylvania 15260, United States.
We report a stereo-differentiating dynamic kinetic asymmetric Rh(I)-catalyzed Pauson-Khand reaction, which provides access to an array of thapsigargin stereoisomers. Using catalyst-control, a consistent stereochemical outcome is achieved at C2─for both matched and mismatched cases─regardless of the allene-yne C8 stereochemistry. The stereochemical configuration for all stereoisomers was assigned by comparing experimental vibrational circular dichroism (VCD) and C NMR to DFT-computed spectra.
View Article and Find Full Text PDFEnhancing the Ground Truth Disparity by MAP Estimation for Developing a Neural-Net Based Stereoscopic Camera.
Sensors (Basel)
December 2024
Department of Information and Communication Engineering, Korea University of Technology and Education (KOREATECH), Cheonan-si 31253, Republic of Korea.
This paper presents a novel method to enhance ground truth disparity maps generated by Semi-Global Matching (SGM) using Maximum a Posteriori (MAP) estimation. SGM, while not producing visually appealing outputs like neural networks, offers high disparity accuracy in valid regions and avoids the generalization issues often encountered with neural network-based disparity estimation. However, SGM struggles with occlusions and textureless areas, leading to invalid disparity values.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!