Artistic style transfer is usually performed between two images, a style image and a content image. Recently, a model named CLIPStyler demonstrated that a natural language description of style could replace the necessity of a reference style image. They achieved this by taking advantage of the CLIP model, which can compute the similarity between a text phrase and an image. In this work, we demonstrate how combining CLIPStyler with a pre-trained, purely vision-based style transfer model can significantly reduce the inference time of CLIPStyler. We call this model FastCLIPStyler. We do a qualitative exploration of the stylised images from both models and argue that our model also has merits in terms of the visual aesthetics of the generated images. Finally, we also point out how FastCLIPStyler can be used to further extend this line of research to create a generalised text-to-style model that does not require optimisation at inference time, which both CLIPStyler and FastCLIPStyler do currently.
The core principle of Variational Inference (VI) is to convert the statistical inference problem of computing complex posterior probability densities into a tractable optimization problem. This property enables VI to be faster than several sampling-based techniques. However, the traditional VI algorithm is not scalable to large data sets and is unable to readily infer out-of-bounds data points without re-running the optimization process. Recent developments in the field, like stochastic-, black box- and amortized-VI, have helped address these issues. Generative modeling tasks nowadays widely make use of amortized VI for its efficiency and scalability, as it utilizes a parameterized function to learn the approximate posterior density parameters. With this paper, we review the mathematical foundations of various VI techniques to form the basis for understanding amortized VI. Additionally, we provide an overview of the recent trends that address several issues of amortized VI, such as the amortization gap, generalization issues, inconsistent representation learning, and posterior collapse. Finally, we analyze alternate divergence measures that improve VI optimization.
During the SARS-Cov-2 pandemic, mask-wearing became an effective tool to prevent spreading and contracting the virus. The ability to monitor the mask-wearing rate in the population would be useful for determining public health strategies against the virus. However, artificial intelligence technologies for detecting face masks have not been deployed at a large scale in real-life to measure the mask-wearing rate in public. In this paper, we present a two-step face mask detection approach consisting of two separate modules: 1) face detection and alignment and 2) face mask classification. This approach allowed us to experiment with different combinations of face detection and face mask classification modules. More specifically, we experimented with PyramidKey and RetinaFace as face detectors while maintaining a lightweight backbone for the face mask classification module. Moreover, we also provide a relabeled annotation of the test set of the AIZOO dataset, where we rectified the incorrect labels for some face images. The evaluation results on the AIZOO and Moxa 3K datasets showed that the proposed face mask detection pipeline surpassed the state-of-the-art methods. The proposed pipeline also yielded a higher mAP on the relabeled test set of the AIZOO dataset than the original test set. Since we trained the proposed model using in-the-wild face images, we can successfully deploy our model to monitor the mask-wearing rate using public CCTV images.
This paper presents a novel Transformer-based facial landmark localization network named Localization Transformer (LOTR). The proposed framework is a direct coordinate regression approach leveraging a Transformer network to better utilize the spatial information in the feature map. An LOTR model consists of three main modules: 1) a visual backbone that converts an input image into a feature map, 2) a Transformer module that improves the feature representation from the visual backbone, and 3) a landmark prediction head that directly predicts the landmark coordinates from the Transformer's representation. Given cropped-and-aligned face images, the proposed LOTR can be trained end-to-end without requiring any post-processing steps. This paper also introduces the smooth-Wing loss function, which addresses the gradient discontinuity of the Wing loss, leading to better convergence than standard loss functions such as L1, L2, and Wing loss. Experimental results on the JD landmark dataset provided by the First Grand Challenge of 106-Point Facial Landmark Localization indicate the superiority of LOTR over the existing methods on the leaderboard and two recent heatmap-based approaches.
Approximating complex probability densities is a core problem in modern statistics. In this paper, we introduce the concept of Variational Inference (VI), a popular method in machine learning that uses optimization techniques to estimate complex probability densities. This property allows VI to converge faster than classical methods, such as, Markov Chain Monte Carlo sampling. Conceptually, VI works by choosing a family of probability density functions and then finding the one closest to the actual probability density -- often using the Kullback-Leibler (KL) divergence as the optimization metric. We introduce the Evidence Lower Bound to tractably compute the approximated probability density and we review the ideas behind mean-field variational inference. Finally, we discuss the applications of VI to variational auto-encoders (VAE) and VAE-Generative Adversarial Network (VAE-GAN). With this paper, we aim to explain the concept of VI and assist in future research with this approach.
Accurate face detection and facial landmark localization are crucial to any face recognition system. We present a series of three single-stage RCNNs with different sized backbones (MobileNetV2-25, MobileNetV2-100, and ResNet101) and a six-layer feature pyramid trained exclusively on the WIDER FACE dataset. We compare the face detection and landmark accuracies using eight context module architectures, four proposed by previous research and four modified versions. We find no evidence that any of the proposed architectures significantly overperform and postulate that the random initialization of the additional layers is at least of equal importance. To show this we present a model that achieves near state-of-the-art performance on WIDER FACE and also provides high accuracy landmarks with a simple context module. We also present results using MobileNetV2 backbones, which achieve over 90% average precision on the WIDER FACE hard validation set while being able to run in real-time. By comparing to other authors, we show that our models exceed the state-of-the-art for similar-sized RCNNs and match the performance of much heavier networks.