论文阅读 [TPAMI-2022] Towards Age-Invariant Face Recognition
论文阅读 [TPAMI-2022] Towards Age-Invariant Face Recognition
论文搜索(studyai.com)
搜索论文: Towards Age-Invariant Face Recognition
搜索论文: http://www.studyai.com/search/whole-site/?q=Towards+Age-Invariant+Face+Recognition
关键字(Keywords)
Face; Face recognition; Aging; Benchmark testing; Training; Feature extraction; Robustness; Age-invariant face recognition; age-invariant model; generative adversarial networks; benchmark dataset
机器视觉
生成对抗; 人脸识别; 人脸生成; 室外场景
摘要(Abstract)
Despite the remarkable progress in face recognition related technologies, reliably recognizing faces across ages remains a big challenge.
尽管人脸识别相关技术取得了显著进步，但跨年龄段可靠地识别人脸仍然是一个巨大的挑战。.
The appearance of a human face changes substantially over time, resulting in significant intra-class variations.
随着时间的推移，人脸的外观会发生显著变化，从而导致显著的阶级内部差异。.
As opposed to current techniques for age-invariant face recognition, which either directly extract age-invariant features for recognition, or first synthesize a face that matches target age before feature extraction, we argue that it is more desirable to perform both tasks jointly so that they can leverage each other.
目前的年龄不变人脸识别技术要么直接提取年龄不变的特征进行识别，要么在特征提取之前首先合成与目标年龄匹配的人脸。与此相反，我们认为更可取的做法是联合执行这两项任务，以便它们能够相互利用。.
To this end, we propose a deep Age-Invariant Model (AIM) for face recognition in the wild with three distinct novelties.
为此，我们提出了一种用于野外人脸识别的深度年龄不变模型（AIM），该模型具有三种不同的新颖性。.
First, AIM presents a novel unified deep architecture jointly performing cross-age face synthesis and recognition in a mutual boosting way.
首先，AIM提出了一种新的统一的深层结构，以相互促进的方式联合执行跨年龄人脸合成和识别。.
Second, AIM achieves continuous face rejuvenation/aging with remarkable photorealistic and identity-preserving properties, avoiding the requirement of paired data and the true age of testing samples.
其次，AIM实现了连续的面部年轻化/老化，具有显著的照片真实感和身份保持特性，避免了成对数据的要求和测试样本的真实年龄。.
Third, effective and novel training strategies are developed for end-to-end learning of the whole deep architecture, which generates powerful age-invariant face representations explicitly disentangled from the age variation.
第三，为整个深层结构的端到端学习开发了有效且新颖的训练策略，该策略可以生成与年龄变化明确分离的强大的年龄不变人脸表示。.
Moreover, we construct a new large-scale Cross-Age Face Recognition (CAFR) benchmark dataset to facilitate existing efforts and push the frontiers of age-invariant face recognition research.
此外，我们还构建了一个新的大规模跨年龄人脸识别（CAFR）基准数据集，以促进现有工作，推动年龄不变人脸识别研究的前沿。.
Extensive experiments on both our CAFR dataset and several other cross-age datasets (MORPH, CACD, and FG-NET) demonstrate the superiority of the proposed AIM model over the state-of-the-arts.
在我们的CAFR数据集和其他几个跨年龄数据集（MORPH、CACD和FG-NET）上进行的大量实验证明了所提出的AIM模型优于现有技术。.
Benchmarking our model on the popular unconstrained face recognition datasets YTF and IJB-C additionally verifies its promising generalization ability in recognizing faces in the wild…
在流行的无约束人脸识别数据集YTF和IJB-C上对我们的模型进行基准测试，进一步验证了该模型在野外人脸识别方面的良好泛化能力。。.
作者(Authors)
[‘Jian Zhao’, ‘Shuicheng Yan’, ‘Jiashi Feng’]