Leveraging Self-Supervision for Cross-Domain Crowd Counting

Liu, Weizhe; Durasov, Nikita; Fua, Pascal

doi:10.1109/cvpr52688.2022.00527

Cited by 32 publications

(11 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods can either explicitly align the feature distributions using a specific distance metric [34,36,53], or implicitly align the distributions using an adversarial loss [8,13,35] or GAN [21,41]. While most current works in domain adaptation focus on image classification [1,2,11,15,24,29,30,37,38,39,46,54], a few have delved into object detection [6,10,16,23,40,45]. Data mixing [64,66] is appealing in the context of UDA because of the opportunity to strategically blend cross-domain information during training.…”

Section: Cross-domain Object Detectionmentioning

confidence: 99%

LossMix: Simplify and Generalize Mixup for Object Detection and Beyond

Vu¹,

Sun²,

Yuan³

et al. 2023

Preprint

View full text Add to dashboard Cite

The success of data mixing augmentations in image classification tasks has been well-received. However, these techniques cannot be readily applied to object detection due to challenges such as spatial misalignment, foreground/background distinction, and plurality of instances. To tackle these issues, we first introduce a novel conceptual framework called Supervision Interpolation, which offers a fresh perspective on interpolation-based augmentations by relaxing and generalizing Mixup. Building on this framework, we propose LossMix, a simple yet versatile and effective regularization that enhances the performance and robustness of object detectors and more. Our key insight is that we can effectively regularize the training on mixed data by interpolating their loss errors instead of ground truth labels. Empirical results on the PASCAL VOC and MS COCO datasets demonstrate that LossMix consistently outperforms currently popular mixing strategies. Furthermore, we design a two-stage domain mixing method that leverages LossMix to surpass Adaptive Teacher (CVPR 2022) and set a new state of the art for unsupervised domain adaptation.* Work done during a residency with Mineral.

show abstract

Section: Cross-domain Object Detectionmentioning

confidence: 99%

LossMix: Simplify and Generalize Mixup for Object Detection and Beyond

Vu¹,

Sun²,

Yuan³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Their approach to the problem of synthetic domain adaptation involved using SE Cycle GAN, a method typically used for style transfer in images, to transform the synthetic images such that they appear in the style of the target domain. Liu et al [103] approach the problem by jointly training a neural network on the synthetic data in a fully-supervised fashion and on the real data in a self-supervised fashion. Their network attempts to solve a proxy task, which involves predicting whether or not an image from the real distribution has been flipped upside down.…”

Section: Synthetic Datamentioning

confidence: 99%

“…GCC only tackles crowd counting problems, likely due to simple 3D assets for humans being highly accessible. Further, there is still a large domain gap between these synthetic images and real images, which has prevented performance gains [103]. Generating high-quality and realistic synthetic images without relying on 3D assets would allow the synthetic counting problem to easily extend to other object classes, and would significantly reduce the annotation burden.…”

Section: Synthetic Datamentioning

confidence: 99%

Counting Objects in Images using DeepLearning: Methods and Current Challenges

D'Alessandro

Mahdavi-Amiri

Hamarneh

2023

Preprint

View full text Add to dashboard Cite

Object counting is an important computer vision application and research topic, which typically involves enumerating the number of objects in an image. Methodologies spanning a broad set of strategies have been proposed for solving object counting problems. These methods have seen an increase in relevance with the recent emergence of several highly successful deep learning techniques, which have led to significant performance improvements on a growing number of annotated counting benchmark datasets. However, despite the recent advancements in deep learning and computer vision, object counting remains a challenging problem with several open research directions. Datasets often contain objects that are highly occluded and which occur across a range of scales and perspectives. Further, popular annotation strategies, like density map annotations, suffer from annotator noise and inconsistency, which creates a performance bottleneck. These annotation strategies also have a high annotation burden, which leads to datasets that are very small when compared to common benchmark datasets in domains like image classification. Given both the significant progress and continued challenges of object counting, this task continues to be an interesting and ongoing research problem. This overview explores the historical context of object counting methods, the fundamental methodologies driving progress, the state of the art methods, and the significant open problems. In particular, we focus on recent trends that attempt to alleviate the problem of the annotation burden for object counting problems.

show abstract

“…The former method [28] is mainly based on CycleGAN [29], which first translates the synthesised labelled data into images with similar styles to the target domain and then trains the translated images in the crowd counting task. The latter method [30][31][32] utilises the adversarial learning strategy of GANs, where a discriminator distinguishes the density maps of the source and target domains produced by a generator to align the data distribution. In this paper, we combine the above two strategies.…”

Section: Domain Adaptationmentioning

confidence: 99%

Domain adaptive crowd counting via dynamic scale aggregation network

Huo

Wang

Qiao

et al. 2023

IET Computer Vision

View full text Add to dashboard Cite

Crowd counting is an important research topic in computer vision. Its goal is to estimate the people's number in an image. Researchers have dramatically improved counting accuracy in recent years by regressing density maps. However, because of the inherent domain shift, the model trained on an expensive manually labelled dataset (source domain) does not perform well on a dataset with scarce labels (target domain). For this issue, a novel dynamic scale aggregation network (DSANet) is proposed to reduce the gaps in style and cross‐domain head scale variations. Specifically, a practical style transfer layer is introduced to reduce the appearance discrepancy between the source and target domains. Then, the translated source and target domain samples are encoded by a generator consisting of the VGG16 network and the dynamic scale aggregation modules (DSA Modules) and produce corresponding density maps. The DSA module can adaptively adjust parameters according to the input features and effectively fuse multi‐scale information to overcome the cross‐domain head scale variations. Next, a discriminator judges the input density map from the source or target domain. Last, domain distributions are aligned through adversarial between the generator and the discriminator. The experiments show that our network outperforms the current state‐of‐the‐art methods and can improve the target domain's performance while maintaining the source domain's performance without significant degradation.

show abstract

Leveraging Self-Supervision for Cross-Domain Crowd Counting

Cited by 32 publications

References 64 publications

LossMix: Simplify and Generalize Mixup for Object Detection and Beyond

LossMix: Simplify and Generalize Mixup for Object Detection and Beyond

Counting Objects in Images using DeepLearning: Methods and Current Challenges

Domain adaptive crowd counting via dynamic scale aggregation network

Contact Info

Product

Resources

About