| 1. |
- Bešenić, Krešimir, et al.
(författare)
-
Picking out the bad apples : unsupervised biometric data filtering for refined age estimation
- 2023
-
Ingår i: The Visual Computer. - Heidelberg, Germany : Springer. - 0178-2789 .- 1432-2315. ; 39, s. 219-237
-
Tidskriftsartikel (refereegranskat)abstract
- Introduction of large training datasets was essential for the recent advancement and success of deep learning methods. Due to the difficulties related to biometric data collection, facial image datasets with biometric trait labels are scarce and usually limited in terms of size and sample diversity. Web-scraping approaches for automatic data collection can produce large amounts of weakly labeled and noisy data. This work is focused on picking out the bad apples from web-scraped facial datasets by automatically removing erroneous samples that impair their usability. The unsupervised facial biometric data filtering method presented in this work greatly reduces label noise levels in web-scraped facial biometric data. Experiments on two large state-of-the-art web-scraped datasets demonstrate the effectiveness of the proposed method with respect to real and apparent age estimation based on five different age estimation methods. Furthermore, we apply the proposed method, together with a newly devised strategy for merging multiple datasets, to data collected from three major web-based data sources (i.e., IMDb, Wikipedia, Google) and derive the new Biometrically Filtered Famous Figure Dataset or B3FD. The proposed dataset, which is made publicly available, enables considerable performance gains for all tested age estimation methods and age estimation tasks. This work highlights the importance of training data quality compared to data quantity and selection of the estimation method.
|
|
| 2. |
- Gogic, Ivan, et al.
(författare)
-
Fast facial expression recognition using local binary features and shallow neural networks
- 2020
-
Ingår i: The Visual Computer. - : SPRINGER. - 0178-2789 .- 1432-2315. ; 36:1, s. 97-112
-
Tidskriftsartikel (refereegranskat)abstract
- Facial expression recognition applications demand accurate and fast algorithms that can run in real time on platforms with limited computational resources. We propose an algorithm that bridges the gap between precise but slow methods and fast but less precise methods. The algorithm combines gentle boost decision trees and neural networks. The gentle boost decision trees are trained to extract highly discriminative feature vectors (local binary features) for each basic facial expression around distinct facial landmark points. These sparse binary features are concatenated and used to jointly optimize facial expression recognition through a shallow neural network architecture. The joint optimization improves the recognition rates of difficult expressions such as fear and sadness. Furthermore, extensive experiments in both within- and cross-database scenarios have been conducted on relevant benchmark data sets for facial expression recognition: CK+, MMI, JAFFE, and SFEW 2.0. The proposed method (LBF-NN) compares favorably with state-of-the-art algorithms while achieving an order of magnitude improvement in execution time.
|
|
| 3. |
- Gogic, Ivan, et al.
(författare)
-
Regression-based methods for face alignment: A survey
- 2021
-
Ingår i: Signal Processing. - : ELSEVIER. - 0165-1684 .- 1872-7557. ; 178
-
Forskningsöversikt (refereegranskat)abstract
- Face alignment is the process of determining a face shape given its location and size in an image. It is used as a basis for other facial analysis tasks and for human-machine interaction and augmented reality applications. It is a challenging problem due to the extremely high variability in facial appearance affected by many external (illumination, occlusion, head pose) and internal factors (race, facial expression). However, advances in deep learning combined with domain-related knowledge from previous research recently demonstrated impressive results nearly saturating the unconstrained benchmark data sets. The focus is shifting towards reducing the computational burden of the face alignment models since real-time performance is required for such a highly dynamic task. Furthermore, many applications target devices on the edge with limited computational power which puts even greater emphasis on computational efficiency. We present the latest development in regression-based approaches that have led towards nearly solving the face alignment problem in an unconstrained scenario. Various regression architectures are systematically explored and recent training techniques discussed in the context of face alignment. Finally, a benchmark comparison of the most successful methods is presented, taking into account execution time as well, to provide a comprehensive overview of this dynamic research field. (C) 2020 Elsevier B.V. All rights reserved.
|
|
| 4. |
- Markus, Nenad, et al.
(författare)
-
Eye pupil localization with an ensemble of randomized trees
- 2014
-
Ingår i: Pattern Recognition. - : Elsevier. - 0031-3203 .- 1873-5142. ; 47:2, s. 578-587
-
Tidskriftsartikel (refereegranskat)abstract
- We describe a method for eye pupil localization based on an ensemble of randomized regression trees and use several publicly available datasets for its quantitative and qualitative evaluation. The method compares well with reported state-of-the-art and runs in real-time on hardware with limited processing power, such as mobile devices.
|
|
| 5. |
- Markus, Nenad, et al.
(författare)
-
Learning Local Descriptors by Optimizing the Keypoint-Correspondence Criterion
- 2016
-
Ingår i: 2016 23RD INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR). - : IEEE COMPUTER SOC. - 9781509048472 ; , s. 2380-2385
-
Konferensbidrag (refereegranskat)abstract
- Current best local descriptors are learned on a large dataset of matching and non-matching keypoint pairs. However, data of this kind is not always available since detailed keypoint correspondences can be hard to establish. On the other hand, we can often obtain labels for pairs of keypoint bags. For example, keypoint bags extracted from two images of the same object under different views form a matching pair, and keypoint bags extracted from images of different objects form a non-matching pair. On average, matching pairs should contain more corresponding keypoints than non-matching pairs. We describe an end-to-end differentiable architecture that enables the learning of local keypoint descriptors from such weakly-labeled data.
|
|
| 6. |
- Markus, Nenad, et al.
(författare)
-
Learning Local Descriptors by Optimizing the Keypoint-Correspondence Criterion: Applications to Face Matching, Learning From Unlabeled Videos and 3D-Shape Retrieval
- 2019
-
Ingår i: IEEE Transactions on Image Processing. - : IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC. - 1057-7149 .- 1941-0042. ; 28:1, s. 279-290
-
Tidskriftsartikel (refereegranskat)abstract
- Current best local descriptors are learned on a large data set of matching and non-matching keypoint pairs. However, data of this kind are not always available, since the detailed keypoint correspondences can be hard to establish. On the other hand, we can often obtain labels for pairs of keypoint bags. For example, keypoint bags extracted from two images of the same object under different views form a matching pair, and keypoint bags extracted from images of different objects form a non-matching pair. On average, matching pairs should contain more corresponding keypoints than non-matching pairs. We describe an end-to-end differentiable architecture that enables the learning of local keypoint descriptors from such weakly labeled data. In addition, we discuss how to improve the method by incorporating the procedure of mining hard negatives. We also show how our approach can be used to learn convolutional features from unlabeled video signals and 3D models.
|
|
| 7. |
- Pandzic, Igor S., et al.
(författare)
-
Faces Everywhere : Towards Ubiquitous Production and Delivery of Face Animation
- 2003
-
Ingår i: MUM 2003. Proceedings of the 2nd International Conference on Mobile and Ubiquitous Multimedia, 10–12 December, 2003, Norrköping, Sweden. - Linköping : Linköping University Electronic Press. - 1581138261 ; , s. 49-56
-
Konferensbidrag (refereegranskat)abstract
- While face animation is still considered one of the toughesttasks in computer animation, its potential application range israpidly moving from the classical field of film production intogames, communications, news delivery and commerce. Tosupport such novel applications, it is important to enableproduction and delivery of face animation on a wide range ofplatforms, from high-end animation systems to the web, gameconsoles and mobile phones. Our goal is to offer a frameworkof tools interconnected by standard formats and protocols andcapable of supporting any imaginable application involvingface animation with the desired level of animation quality,automatic production wherever it is possible, and delivery ona wide range of platforms. While this is clearly an ongoingtask, we present the current state of development along withseveral case studies showing that a wide range of applicationsis already enabled.
|
|
| 8. |
- Zoric, Goranka, et al.
(författare)
-
On creating multimodal virtual humans-real time speech driven facial gesturing
- 2011
-
Ingår i: MULTIMEDIA TOOLS AND APPLICATIONS. - : Springer Science Business Media. - 1380-7501 .- 1573-7721. ; 54:1, s. 165-179
-
Tidskriftsartikel (refereegranskat)abstract
- Because of extensive use of different computer devices, human-computer interaction design nowadays moves towards creating user centric interfaces. It assumes incorporating different modalities that humans use in everyday communication. Virtual humans, who look and behave believably, fit perfectly in the concept of designing interfaces in more natural, effective, as well as social oriented way. In this paper we present a novel method for automatic speech driven facial gesturing for virtual humans capable of real time performance. Facial gestures included are various nods and head movements, blinks, eyebrow gestures and gaze. A mapping from speech to facial gestures is based on the prosodic information obtained from the speech signal. It is realized using a hybrid approach-Hidden Markov Models, rules and global statistics. Further, we test the method using an application prototype-a system for speech driven facial gesturing suitable for virtual presenters. Subjective evaluation of the system confirmed that the synthesized facial movements are consistent and time aligned with the underlying speech, and thus provide natural behavior of the whole face.
|
|
