Optimizing Filter Size in Convolutional Neural Networks for Facial Action Unit Recognition

Han, Shizhong; Meng, Zibo; Li, Zhiyuan; O’Reilly, James T.; Cai, Jie; Wang, Xiaofeng; Tong, Yan

doi:10.1109/cvpr.2018.00532

Cited by 90 publications

(75 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Very recently, Han et al [52] propose to use boosting to select discriminative neurons for facial action unit classification. They employ decision stumps on top of single neurons as weak learners, and learn weighting factors for each of these neurons by offline AdaBoost [42] applied to each mini-batch separately.…”

Section: Boosting For Cnnsmentioning

confidence: 99%

Deep Metric Learning with BIER: Boosting Independent Embeddings Robustly

Opitz

Waltner

Possegger

et al. 2020

IEEE Trans. Pattern Anal. Mach. Intell.

133

View full text Add to dashboard Cite

Learning similarity functions between image pairs with deep neural networks yields highly correlated activations of embeddings. In this work, we show how to improve the robustness of such embeddings by exploiting the independence within ensembles. To this end, we divide the last embedding layer of a deep network into an embedding ensemble and formulate the task of training this ensemble as an online gradient boosting problem. Each learner receives a reweighted training sample from the previous learners. Further, we propose two loss functions which increase the diversity in our ensemble. These loss functions can be applied either for weight initialization or during training. Together, our contributions leverage large embedding sizes more effectively by significantly reducing correlation of the embedding and consequently increase retrieval accuracy of the embedding. Our method works with any differentiable loss function and does not introduce any additional parameters during test time. We evaluate our metric learning method on image retrieval tasks and show that it improves over state-of-the-art methods on the CUB-200-2011, Cars-196, Stanford Online Products, In-Shop Clothes Retrieval and VehicleID datasets. Therefore, our findings suggest that by dividing deep networks at the end into several smaller and diverse networks, we can significantly reduce overfitting.

show abstract

Section: Boosting For Cnnsmentioning

confidence: 99%

Deep Metric Learning with BIER: Boosting Independent Embeddings Robustly

Opitz

Waltner

Possegger

et al. 2020

IEEE Trans. Pattern Anal. Mach. Intell.

133

View full text Add to dashboard Cite

show abstract

“…While not yet particularly popular, research efforts have already been devoted to investigate the advantages of such irregular CNNs [53]. Our HCNN would facilitate such possible future research endeavors more friendly.…”

Section: Discussionmentioning

confidence: 99%

H-CNN: Spatial Hashing Based CNN for 3D Shape Analysis

Shao

Yang

Weng

et al. 2020

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

We present a novel spatial hashing based data structure to facilitate 3D shape analysis using convolutional neural networks (CNNs). Our method well utilizes the sparse occupancy of 3D shape boundary and builds hierarchical hash tables for an input model under different resolutions. Based on this data structure, we design two efficient GPU algorithms namely hash2col and col2hash so that the CNN operations like convolution and pooling can be efficiently parallelized. The spatial hashing is nearly minimal, and our data structure is almost of the same size as the raw input. Compared with state-of-the-art octree-based methods, our data structure significantly reduces the memory footprint during the CNN training. As the input geometry features are more compactly packed, CNN operations also run faster with our data structure. The experiment shows that, under the same network structure, our method yields comparable or better benchmarks compared to the state-of-the-art while it has only one-third memory consumption. Such superior memory performance allows the CNN to handle high-resolution shape analysis.

show abstract

“…Although these results are not directly comparable because of different experimental settings, they indicate that our method trained with labels at sequence-level and a small portion of labelled frames can still show competitive performance. It is known that Supervised Deep Learning models require a large number of samples to be effectively trained [49]. Thus, this still limits their application to Facial Behavior Analysis, where the annotation process is laborious and labelled data is scarce.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

Multi-Instance Dynamic Ordinal Random Fields for Weakly Supervised Facial Behavior Analysis

Ruiz

Rudovic

Binefa

et al. 2018

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

We propose a Multi-Instance-Learning (MIL) approach for weakly-supervised learning problems, where a training set is formed by bags (sets of feature vectors or instances) and only labels at bag-level are provided. Specifically, we consider the Multi-Instance Dynamic-Ordinal-Regression (MI-DOR) setting, where the instance labels are naturally represented as ordinal variables and bags are structured as temporal sequences. To this end, we propose Multi-Instance Dynamic Ordinal Random Fields (MI-DORF). In this framework, we treat instance-labels as temporally-dependent latent variables in an Undirected Graphical Model. Different MIL assumptions are modelled via newly introduced high-order potentials relating bag and instance-labels within the energy function of the model. We also extend our framework to address the Partially-Observed MI-DOR problems, where a subset of instance labels are available during training.We show on the tasks of weakly-supervised facial behavior analysis, Facial Action Unit (DISFA dataset) and Pain (UNBC dataset) Intensity estimation, that the proposed framework outperforms alternative learning approaches. Furthermore, we show that MIDORF can be employed to reduce the data annotation efforts in this context by large-scale.

show abstract

Optimizing Filter Size in Convolutional Neural Networks for Facial Action Unit Recognition

Cited by 90 publications

References 38 publications

Deep Metric Learning with BIER: Boosting Independent Embeddings Robustly

Deep Metric Learning with BIER: Boosting Independent Embeddings Robustly

H-CNN: Spatial Hashing Based CNN for 3D Shape Analysis

Multi-Instance Dynamic Ordinal Random Fields for Weakly Supervised Facial Behavior Analysis

Contact Info

Product

Resources

About