Face Recognition Using Synthetic Face Data

In the field of deep learning applied to face recognition, securing large-scale, high-quality datasets is vital for attaining precise and reliable results. However, amassing significant volumes of high-quality real data faces hurdles such as time limitations, financial burdens, and privacy issues. Furthermore, prevalent datasets are often impaired by racial biases and annotation inaccuracies. In this paper, we underscore the promising application of synthetic data, generated through rendering digital faces via our computer graphics pipeline, in achieving competitive results with the state-of-the-art on synthetic data across multiple benchmark datasets. By finetuning the model,we obtain results that rival those achieved when training with hundreds of thousands of real images (98.7% on LFW [1]). We further investigate the contribution of adding intra-class variance factors (e.g., makeup, accessories, haircuts) on model performance. Finally, we reveal the sensitivity of pre-trained face recognition models to alternating specific parts of the face by leveraging the granular control capability in our platform.

Knowing the Distance: Understanding the Gap Between Synthetic and Real Data For Face Parsing

The use of synthetic data for training computer vision algorithms has become increasingly popular due to its cost-effectiveness, scalability, and ability to provide accurate multi-modality labels. Although recent studies have demonstrated impressive results when training networks solely on synthetic data, there remains a performance gap between synthetic and real data that is commonly attributed to lack of photorealism. The aim of this study is to investigate the gap in greater detail for the face parsing task. We differentiate between three types of gaps: distribution gap, label gap, and photorealism gap. Our findings show that the distribution gap is the largest contributor to the performance gap, accounting for over 50% of the gap. By addressing this gap and accounting for the labels gap, we demonstrate that a model trained on synthetic data achieves comparable results to one trained on a similar amount of real data. This suggests that synthetic data is a viable alternative to real data, especially when real data is limited or difficult to obtain. Our study highlights the importance of content diversity in synthetic datasets and challenges the notion that the photorealism gap is the most critical factor affecting the performance of computer vision models trained on synthetic data.

Why You Should Try the Real Data for the Scene Text Recognition

Recent works in the text recognition area have pushed forward the recognition results to the new horizons. But for a long time a lack of large human-labeled natural text recognition datasets has been forcing researchers to use synthetic data for training text recognition models. Even though synthetic datasets are very large (MJSynth and SynthTest, two most famous synthetic datasets, have several million images each), their diversity could be insufficient, compared to natural datasets like ICDAR and others. Fortunately, the recently released text-recognition annotation for OpenImages V5 dataset has comparable with synthetic dataset number of instances and more diverse examples. We have used this annotation with a Text Recognition head architecture from the Yet Another Mask Text Spotter and got comparable to the SOTA results. On some datasets we have even outperformed previous SOTA models. In this paper we also introduce a text recognition model. The model's code is available.