Few-Shot Adaptive Gaze Estimation

Park, Seonwook; Mello, Shalini De; Molchanov, Pavlo; Iqbal, Umar; Hilliges, Otmar; Kautz, Jan

doi:10.1109/iccv.2019.00946

Cited by 169 publications

(142 citation statements)

References 50 publications

Supporting

Mentioning

128

Contrasting

Order By: Relevance

“…There are many methods to estimate gaze direction using the images of the eyes. Most accurate methods use neural networks (described in [77] and [46]) but for rough gaze direction estimation, it is enough to apply simple computer vision algorithms such as pupil position detection [78].…”

Section: B Distraction Detectionmentioning

confidence: 99%

Driver Distraction Detection Methods: A Literature Review and Framework

et al. 2021

View full text Add to dashboard Cite

show abstract

Section: B Distraction Detectionmentioning

confidence: 99%

Driver Distraction Detection Methods: A Literature Review and Framework

et al. 2021

View full text Add to dashboard Cite

show abstract

“…For the user-specific pupil adaptation, Yu et al [ 19 ] generated additional training samples through the synthesis of gaze-redirected eye images from existing reference samples. Similar to [ 19 , 20 ] also proposed a framework for a few-shot adaptive gaze estimation for the learning of person-specific gaze networks by applying very few calibration samples. However, these GAN user-specific gaze adaptation approaches primarily use a transforming GAN and encoder-decoder architectures, which require fine-tuning to adapt the model to a new subject, and thus user-specificity and personalization are computationally intensive, requiring a large amount of calibration data, and cannot be run on low-spec devices [ 1 ].…”

Section: Related Workmentioning

confidence: 99%

“…State-of-the-art DNN-based studies [ 20 , 21 ] follow similar steps in a single network using a layer-by-layer structure and an end-to-end learning paradigm. Although a DNN-based pupil estimation has achieved outstanding results, it has certain limitations, such as too many hyper-parameters, a reliance on black-box training, high processing costs, and the requirement of vast amounts of training data.…”

Section: Rule Distillation Of Cascadementioning

confidence: 99%

Energy Efficient Pupil Tracking Based on Rule Distillation of Cascade Regression Forest

Kim

Jeong

2020

Sensors

View full text Add to dashboard Cite

As the demand for human-friendly computing increases, research on pupil tracking to facilitate human–machine interactions (HCIs) is being actively conducted. Several successful pupil tracking approaches have been developed using images and a deep neural network (DNN). However, common DNN-based methods not only require tremendous computing power and energy consumption for learning and prediction; they also have a demerit in that an interpretation is impossible because a black-box model with an unknown prediction process is applied. In this study, we propose a lightweight pupil tracking algorithm for on-device machine learning (ML) using a fast and accurate cascade deep regression forest (RF) instead of a DNN. Pupil estimation is applied in a coarse-to-fine manner in a layer-by-layer RF structure, and each RF is simplified using the proposed rule distillation algorithm for removing unimportant rules constituting the RF. The goal of the proposed algorithm is to produce a more transparent and adoptable model for application to on-device ML systems, while maintaining a precise pupil tracking performance. Our proposed method experimentally achieves an outstanding speed, a reduction in the number of parameters, and a better pupil tracking performance compared to several other state-of-the-art methods using only a CPU.

show abstract

“…On the other hand, existing object (or rigid body) keypoint localization approaches, e.g. [19,29,22,18], always fail to successively explore CL among keypoints. Few studies tackled the prob- Figure 1: (a) KeypointNet [22].…”

Section: Introductionmentioning

confidence: 99%

Conditional Link Prediction of Category-Implicit Keypoint Detection

Yi-Ge¹,

Fan²,

Liu³

et al. 2020

Preprint

View full text Add to dashboard Cite

<div>Keypoints of objects reflect their concise abstractions, while the corresponding connection links (CL) build the skeleton by detecting the intrinsic relations between keypoints. Existing approaches are typically computationally-intensive, inapplicable for instances belonging to multiple classes, and/or infeasible to simultaneously encode connection information. To address the aforementioned issues, we propose an end-to-end category-implicit Keypoint and Link Prediction Network (KLPNet), which is the first approach for simultaneous semantic keypoint detection (for multi-class instances) and CL rejuvenation. In our KLPNet, a novel Conditional Link Prediction Graph is proposed for link prediction among keypoints that are contingent on a predefined category. Furthermore, a Cross-stage Keypoint Localization Module (CKLM) is introduced to explore feature aggregation for coarse-to-fine keypoint localization. Comprehensive experiments conducted on three publicly available benchmarks demonstrate that our KLPNet consistently outperforms all other state-of-the-art approaches. Furthermore, the experimental results of CL prediction also show the effectiveness of our KLPNet with respect to occlusion problems.</div>

show abstract

Few-Shot Adaptive Gaze Estimation

Cited by 169 publications

References 50 publications

Driver Distraction Detection Methods: A Literature Review and Framework

Driver Distraction Detection Methods: A Literature Review and Framework

Energy Efficient Pupil Tracking Based on Rule Distillation of Cascade Regression Forest

Conditional Link Prediction of Category-Implicit Keypoint Detection

Contact Info

Product

Resources

About