ForgeryNet: A Versatile Benchmark for Comprehensive Forgery Analysis

He, Yakun; Gan, Bei; Chen, Siyu; Zhou, Yichun; Yin, Guojun; Song, Luchuan; Sheng, Lu; Shao, Jing; Liu, Ziwei

doi:10.1109/cvpr46437.2021.00434

Cited by 81 publications

(23 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The original test set is denoted as "full set" in the experiments. Unlike temporal action localization methods [42,46] that are using only average precision, we follow the protocol proposed in ForgeryNet [23] and use both average precision (AP) and average recall (AR) as the evaluation metrics for the quantitative comparison. For AP, we follow the protocol of ActivityNet [7] to set the IoU thresholds to 0.5, 0.75 and 0.95.…”

Section: Methodsmentioning

confidence: 99%

“…Recently, FakeAVCeleb [31] was released focusing on both face-swap and face-reenactment methods with manipulated audio and video. ForgeryNet [23] is the latest contribution to the growing list of deepfake detection datasets. This large-scale dataset is also centered around video-only identity manipulation and is suitable for the tasks of video/image classification and spatial/temporal forgery localization.…”

Section: Related Workmentioning

confidence: 99%

“…To the best of our knowledge, all these previous datasets assume that face manipulation occurs in most of the frames of the video [11]. Only the latest one, ForgeryNet [23], provides examples of the important problem of temporal forgery localization since it includes random face swapping applied to parts of some videos. However, the manipulations present in that dataset are only identity modifications that do not necessarily alter the meaning of the content.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Do You Really Mean That? Content Driven Audio-Visual Deepfake Dataset and Multimodal Method for Temporal Forgery Localization

Cai¹,

Stefanov²,

Dhall³

et al. 2022

Preprint

View full text Add to dashboard Cite

Due to its high societal impact, deepfake detection is getting active attention in the computer vision community. Most deepfake detection methods rely on identity, facial attribute and adversarial perturbation based spatio-temporal modifications at the whole video or random locations, while keeping the meaning of the content intact. However, a sophisticated deepfake may contain only a small segment of video/audio manipulation, through which the meaning of the content can be, for example, completely inverted from sentiment perspective. To address this gap, we introduce a content driven audio-visual deepfake dataset, termed as Localized Audio Visual DeepFake (LAV-DF), explicitly designed for the task of learning temporal forgery localization. Specifically, the content driven audio-visual manipulations are performed at strategic locations in order to change the sentiment polarity of the whole video. Our baseline method for benchmarking the proposed dataset is a 3DCNN model, termed as Boundary Aware Temporal Forgery Detection (BA-TFD), which is guided via contrastive, boundary matching and frame classification loss functions. Our extensive quantitative analysis demonstrates the strong performance of the proposed method for both tasks of temporal forgery localization and deepfake detection.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Do You Really Mean That? Content Driven Audio-Visual Deepfake Dataset and Multimodal Method for Temporal Forgery Localization

Cai¹,

Stefanov²,

Dhall³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However the majority of approaches are frame-based, classifying the video frame by frame. In order to train effective deep learning models for deepfake detection, a number of datasets have been created over the years, including the first DF-TIMIT [23], UADFC [35] and FaceForensics++ [29], Celeb-DF [26], Google Deepfake Detection Dataset [10] and the more recent DFDC [9], Deepforensics [16] and ForgeryNet [14]. The latter dataset is the most complete, largest and includes the greater variety of existing deepfake generation methods, since it is still recently published there are not many papers based on it.…”

Section: Deepfake Detectionmentioning

confidence: 99%

“…To be able to validate the ability of a neural network to detect deepfakes generated by methods other than those used for the construction of the training set, it is necessary to use a dataset containing a multitude of deepfake generation methods and keeps track of them. For this reason the dataset selected to carry out the experiments is ForgeryNet [14], one of the widest deepfake datasets available. ForgeryNet consists of 2.9M images and 220k video clips.…”

Section: Datasetmentioning

confidence: 99%

Cross-Forgery Analysis of Vision Transformers and CNNs for Deepfake Image Detection

Coccomini

Caldelli

Falchi

et al. 2022

Proceedings of the 1st International Workshop on Multimedia AI Against Disinformation

View full text Add to dashboard Cite

Deepfake Generation Techniques are evolving at a rapid pace, making it possible to create realistic manipulated images and videos and endangering the serenity of modern society. The continual emergence of new and varied techniques brings with it a further problem to be faced, namely the ability of deepfake detection models to update themselves promptly in order to be able to identify manipulations carried out using even the most recent methods. This is an extremely complex problem to solve, as training a model requires large amounts of data, which are difficult to obtain if the deepfake generation method is too recent. Moreover, continuously retraining a network would be unfeasible. In this paper, we ask ourselves if, among the various deep learning techniques, there is one that is able to generalise the concept of deepfake to such an extent that it does not remain tied to one or more specific deepfake generation methods used in the training set. We compared a Vision Transformer with an EfficientNetV2 on a cross-forgery context based on the ForgeryNet dataset. From our experiments, It emerges that EfficientNetV2 has a greater tendency to specialize often obtaining better results on training methods while Vision Transformers exhibit a superior generalization ability that makes them more competent even on images generated with new methodologies.

show abstract

DeepFakes Detection: the DeeperForensics Dataset and Challenge

Jiang

Qian

et al. 2022

Advances in Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

Recent years have witnessed exciting progress in automatic face swapping and editing. Many techniques have been proposed, facilitating the rapid development of creative content creation. The emergence and easy accessibility of such techniques, however, also cause potential unprecedented ethical and moral issues. To this end, academia and industry proposed several effective forgery detection methods. Nonetheless, challenges could still exist. (1) Current face manipulation advances can produce high-fidelity fake videos, rendering forgery detection challenging. (2) The generalization capability of most existing detection models is poor, particularly in real-world scenarios where the media sources and distortions are unknown. The primary difficulty in overcoming these challenges is the lack of amenable datasets for real-world face forgery detection. Most existing datasets are either of a small number, of low quality, or overly artificial. Meanwhile, the large distribution gap between training data and actual test videos also leads to weak generalization ability. In this chapter, we present our on-going effort of constructing DeeperForensics-1.0, a large-scale forgery detection dataset, to address the challenges above. We discuss approaches to ensure the quality and diversity of the dataset. Besides, we describe the observations we obtained from organizing DeeperForensics Challenge 2020, a real-world face forgery detection competition based on DeeperForensics-1.0. Specifically, we summarize the winning solutions and provide some discussions on potential research directions.

show abstract

ForgeryNet: A Versatile Benchmark for Comprehensive Forgery Analysis

Cited by 81 publications

References 25 publications

Do You Really Mean That? Content Driven Audio-Visual Deepfake Dataset and Multimodal Method for Temporal Forgery Localization

Do You Really Mean That? Content Driven Audio-Visual Deepfake Dataset and Multimodal Method for Temporal Forgery Localization

Cross-Forgery Analysis of Vision Transformers and CNNs for Deepfake Image Detection

DeepFakes Detection: the DeeperForensics Dataset and Challenge

Contact Info

Product

Resources

About