Research map / Image generation / Image restoration
Online Robust Image Alignment via Subspace Learning From Gradient Orientations
Qingqing Zheng, Yi Wang, Pheng-Ann Heng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| align in the gradient orientation domain with online alignment instead of SIFT alignment | SRC face recognition · LFW face recognition (SRC, 20 train / 15 test per subject) | recognition rate (%): 52.28 -> 91.56 | Improved |
| replace t-GRASTA alignment with the proposed IGO online alignment | SRC face recognition · LFW face recognition (SRC, 20 train / 15 test per subject) | recognition rate (%): 81.40 -> 91.56 | Improved |
| replace batch RASL alignment with the proposed IGO online alignment | SRC face recognition · LFW face recognition (SRC, 20 train / 15 test per subject) | recognition rate (%): 88.42 -> 91.56 | Improved |
| use online subspace alignment instead of incrementally rerunning RASL | LFW alignment · LFW alignment (per subject, 35 images) | seconds per image: 19.2 -> 0.7 | Improved |
| keep only a low-dimensional subspace rather than all images as RASL does | Online alignment · 100 images of 100x100 pixels, k=5, Aff(2) | memory vs RASL: 3% of the memory of RASL | Improved |
| compare the proposed method with RASL on alignment accuracy | Online alignment · LFW landmark STD | landmark STD (qualitative) | No clear effect |
| use IGO subspace instead of intensity-based low-rank methods under shadows | Online alignment · LFW with artificial shadows | alignment quality (qualitative) | Improved |
Closest papers by meaning
- Intrinsic Grassmann Averages for Online Linear and Robust Subspace Learning (CVPR 2017 · similarity 0.84)
- PRISE: Demystifying Deep Lucas-Kanade With Strongly Star-Convex Constraints for Multimodel Image Alignment (CVPR 2023 · similarity 0.84)
- Robust Face Alignment via Deep Progressive Reinitialization and Adaptive Error-driven Learning (TPAMI 2021 · similarity 0.84)
- Side Information for Face Completion: A Robust PCA Approach (TPAMI 2019 · similarity 0.84)
- FastJAM: a Fast Joint Alignment Model for Images (NeurIPS 2025 · similarity 0.83)
- Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection (ICML 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
