TFW: Annotated Thermal Faces in the Wild Dataset

Kuzdeuov, Askat; Aubakirova, Dana; Koishigarina, Darina; Varol, Hüseyin Atakan

doi:10.36227/techrxiv.17004538.v1

“…Regarding the TFW dataset, the authors reported 100% for the indoor set and 97.2% AP for the outdoor set as their best results [24]. In our case, the AnyFace model achieved 100% for the indoor set and 99.2% for the outdoor set without test set augmentation.…”

Section: Results On the Test Setmentioning

confidence: 54%

AnyFace: A Data-Centric Approach For Input-Agnostic Face Detection

Kuzdeuov¹,

Koishigarina²,

Varol³

2022

Preprint

0

View full text Add to dashboard Cite

<p>Face detection is a mandatory step in many computer vision applications, such as face recognition, emotion recognition, age detection, virtual makeup, and vital sign monitoring. Thanks to advancements in deep learning and the introduction of annotated large-scale datasets, numerous applications have been developed for human faces. Recently, other domains, such as animals and cartoon characters, have started gaining attention but still lag far behind human faces. The biggest challenge is the limited number of annotated face datasets in these domains. The manual labeling of large-scale datasets is tedious and requires substantial human labor. In this regard, we present an input-agnostic face detector to ease the annotation of various face datasets. We propose a simple but effective data-centric approach instead of building a specific neural network architecture. Specifically, we trained a face detection model, YOLO5Face, on human, animal, and cartoon face datasets. The experiments show that the model can achieve accurate results in all domains. In addition, the model achieved decent results for animals and cartoon characters different from the ones in the training set. This implies that the model can extract agnostic facial features. We have made the source code and pre-trained models publicly available at https://github.com/IS2AI/AnyFace to stimulate research in these fields.</p>

show abstract

“…Therefore, a limited number of face detection models are available for these cameras. For instance, the YOLOv5 model was used to implement a thermal face detection model [24]. The authors collected 9,982 thermal images and manually annotated 16,509 faces to train the model.…”

Section: Related Workmentioning

confidence: 99%

“…The dataset contains 60,000 images (109,807 annotated faces) for the face detection task. TFW dataset [24] Images Faces Wider Face [22] 32,203 393,703 AnimalWeb [34] 19,079 22,451 iCartoonFace [40] 60,000 109,807 TFW [24] 9,982 16,509…”

Section: A Datasetsmentioning

confidence: 99%

AnyFace: A Data-Centric Approach For Input-Agnostic Face Detection

Kuzdeuov¹,

Koishigarina²,

Varol³

2022

Preprint

2

0

View full text Add to dashboard Cite

<p>Face detection is a mandatory step in many computer vision applications, such as face recognition, emotion recognition, age detection, virtual makeup, and vital sign monitoring. Thanks to advancements in deep learning and the introduction of annotated large-scale datasets, numerous applications have been developed for human faces. Recently, other domains, such as animals and cartoon characters, have started gaining attention but still lag far behind human faces. The biggest challenge is the limited number of annotated face datasets in these domains. The manual labeling of large-scale datasets is tedious and requires substantial human labor. In this regard, we present an input-agnostic face detector to ease the annotation of various face datasets. We propose a simple but effective data-centric approach instead of building a specific neural network architecture. Specifically, we trained a face detection model, YOLO5Face, on human, animal, and cartoon face datasets. The experiments show that the model can achieve accurate results in all domains. In addition, the model achieved decent results for animals and cartoon characters different from the ones in the training set. This implies that the model can extract agnostic facial features. We have made the source code and pre-trained models publicly available at https://github.com/IS2AI/AnyFace to stimulate research in these fields.</p>

show abstract