Data Augmentation via Latent Space Interpolation for Image Classification

Liu, Xiaofeng; Zou, Yang; Kong, Lingsheng; Diao, Zhihui; Yan, Junliang; Wang, Jun; Li, Site; Jia, Ping; You, Jane

doi:10.1109/icpr.2018.8545506

Cited by 72 publications

(33 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the previous works only consider the image domain, and the spatiotemporal FER does not significantly outperforms aggregation methods [2]. To the best of our knowledge, this is the first effort to investigate the compressed video FER, which is orthogonal to these advantages and can be easily added to each other [37], [38], [39], [40], [41]. Video compression Usually, the video codecs separates a video into several Group Of Pictures (GOP).…”

Section: Related Workmentioning

confidence: 99%

Identity-aware Facial Expression Recognition in Compressed Video

Liu

Jin

Han

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Identity-aware Facial Expression Recognition in Compressed Video

Liu

Jin

Han

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

“…Unsupervised domain adaptation (UDA) with deep networks [20], [21], [22], [23], [24] targets to learn domain invariant embeddings by minimizing the cross-domain difference of feature distributions with certain criteria [25], [26]. Examples of these methods include maximum mean discrepancy (MMD), deep Correlation Alignment (CORAL), sliced Wasserstein discrepancy, adversarial learning at input-level, feature level, output space level, etc [10], [27].…”

Section: Related Workmentioning

confidence: 99%

Energy-constrained Self-training for Unsupervised Domain Adaptation

Liu

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Unsupervised domain adaptation (UDA) aims to transfer the knowledge on a labeled source domain distribution to perform well on an unlabeled target domain. Recently, the deep self-training involves an iterative process of predicting on the target domain and then taking the confident predictions as hard pseudo-labels for retraining. However, the pseudo-labels are usually unreliable, and easily leading to deviated solutions with propagated errors. In this paper, we resort to the energy-based model and constrain the training of the unlabeled target sample with the energy function minimization objective. It can be applied as a simple additional regularization. In this framework, it is possible to gain the benefits of the energy-based model, while retaining strong discriminative performance following a plugand-play fashion. We deliver extensive experiments on the most popular and large scale UDA benchmarks of image classification as well as semantic segmentation to demonstrate its generality and effectiveness.

show abstract

“…(Long, Shelhamer, and Darrell 2015) introduced a fully convolutional network for pixel or super pixel-wise classification. The conventional approaches usually employ CE loss (Liu et al 2018d;Liu et al 2018e;Liu et al 2019d;Liu et al 2019a), which equally evaluates the errors incurred by all image pixels/classes without taking into account the different severity-level of different mistakes (Chen, Gong, and Yang 2018).…”

Section: Related Work Semantic Segmentationmentioning

confidence: 99%

Importance-Aware Semantic Segmentation in Self-Driving with Discrete Wasserstein Training

Liu

Han

Bai

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

Semantic segmentation (SS) is an important perception manner for self-driving cars and robotics, which classifies each pixel into a pre-determined class. The widely-used cross entropy (CE) loss-based deep networks has achieved significant progress w.r.t. the mean Intersection-over Union (mIoU). However, the cross entropy loss can not take the different importance of each class in an self-driving system into account. For example, pedestrians in the image should be much more important than the surrounding buildings when make a decisions in the driving, so their segmentation results are expected to be as accurate as possible. In this paper, we propose to incorporate the importance-aware inter-class correlation in a Wasserstein training framework by configuring its ground distance matrix. The ground distance matrix can be pre-defined following a priori in a specific task, and the previous importance-ignored methods can be the particular cases. From an optimization perspective, we also extend our ground metric to a linear, convex or concave increasing function w.r.t. pre-defined ground distance. We evaluate our method on CamVid and Cityscapes datasets with different backbones (SegNet, ENet, FCN and Deeplab) in a plug and play fashion. In our extenssive experiments, Wasserstein loss demonstrates superior segmentation performance on the predefined critical classes for safe-driving.

show abstract

Data Augmentation via Latent Space Interpolation for Image Classification

Cited by 72 publications

References 12 publications

Identity-aware Facial Expression Recognition in Compressed Video

Identity-aware Facial Expression Recognition in Compressed Video

Energy-constrained Self-training for Unsupervised Domain Adaptation

Importance-Aware Semantic Segmentation in Self-Driving with Discrete Wasserstein Training

Contact Info

Product

Resources

About