skip to main content


Title: Crowd Counting with Minimal Data Using Generative Adversarial Networks for Multiple Target Regression
In this work, we use a generative adversarial network (GAN) to train crowd counting networks using minimal data. We describe how GAN objectives can be modified to allow for the use of unlabeled data to benefit inference training in semi-supervised learning. More generally, we explain how these same methods can be used in more generic multiple regression target semi-supervised learning, with crowd counting being a demonstrative example. Given a convolutional neural network (CNN) with capabilities equivalent to the discriminator in the GAN, we provide experimental results which show that our GAN is able to outperform the CNN even when the CNN has access to significantly more labeled data. This presents the potential of training such networks to high accuracy with little data. Our primary goal is not to outperform the state-of-the-art using an improved method on the entire dataset, but instead we work to show that through semi-supervised learning we can reduce the data required to train an inference network to a given accuracy. To this end, systematic experiments are performed with various numbers of images and cameras to show under which situations the semi-supervised GANs can improve results.  more » « less
Award ID(s):
1737533 1137172
NSF-PAR ID:
10065141
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
2018 IEEE Winter Conference on Applications of Computer Vision (WACV)
Page Range / eLocation ID:
1151 to 1159
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. In this work, we generalize semi-supervised generative adversarial networks (GANs) from classification problems to regression for use in dense crowd counting. In the last several years, the importance of improving the training of neural networks using semi-supervised training has been thoroughly demonstrated for classification problems. This work presents a dual-goal GAN which seeks both to provide the number of individuals in a densely crowded scene and distinguish between real and generated images. This method allows the dual-goal GAN to benefit from unlabeled data in the training process, improving the predictive capabilities of the discriminating network compared to the fully-supervised version of the network. Typical semi-supervised GANs are unable to function in the regression regime due to biases introduced when using a single prediction goal. Using the proposed approach, the amount of data which needs to be annotated for dense crowd counting can be significantly reduced. 
    more » « less
  2. Phenology––the timing of life-history events––is a key trait for understanding responses of organisms to climate. The digitization and online mobilization of herbarium specimens is rapidly advancing our understanding of plant phenological response to climate and climatic change. The current common practice of manually harvesting data from individual specimens greatly restricts our ability to scale data collection to entire collections. Recent investigations have demonstrated that machine-learning models can facilitate data collection from herbarium specimens. However, present attempts have focused largely on simplistic binary coding of reproductive phenology (e.g., flowering or not). Here, we use crowd-sourced phenological data of numbers of buds, flowers, and fruits of more than 3000 specimens of six common wildflower species of the eastern United States (Anemone canadensis, A. hepatica, A. quinquefolia, Trillium erectum, T. grandiflorum, and T. undulatum} to train a model using Mask R-CNN to segment and count phenological features. A single global model was able to automate the binary coding of reproductive stage with greater than 90% accuracy. Segmenting and counting features were also successful, but accuracy varied with phenological stage and taxon. Counting buds was significantly more accurate than flowers or fruits. Moreover, botanical experts provided more reliable data than either crowd-sourcers or our Mask R-CNN model, highlighting the importance of high-quality human training data. Finally, we also demonstrated the transferability of our model to automated phenophase detection and counting of the three Trillium species, which have large and conspicuously-shaped reproductive organs. These results highlight the promise of our two-phase crowd-sourcing and machine-learning pipeline to segment and count reproductive features of herbarium specimens, providing high-quality data with which to study responses of plants to ongoing climatic change. 
    more » « less
  3. Given earth imagery with spectral features on a terrain surface, this paper studies surface segmentation based on both explanatory features and surface topology. The problem is important in many spatial and spatiotemporal applications such as flood extent mapping in hydrology. The problem is uniquely challenging for several reasons: first, the size of earth imagery on a terrain surface is often much larger than the input of popular deep convolutional neural networks; second, there exists topological structure dependency between pixel classes on the surface, and such dependency can follow an unknown and non-linear distribution; third, there are often limited training labels. Existing methods for earth imagery segmentation often divide the imagery into patches and consider the elevation as an additional feature channel. These methods do not fully incorporate the spatial topological structural constraint within and across surface patches and thus often show poor results, especially when training labels are limited. Existing methods on semi-supervised and unsupervised learning for earth imagery often focus on learning representation without explicitly incorporating surface topology. In contrast, we propose a novel framework that explicitly models the topological skeleton of a terrain surface with a contour tree from computational topology, which is guided by the physical constraint (e.g., water flow direction on terrains). Our framework consists of two neural networks: a convolutional neural network (CNN) to learn spatial contextual features on a 2D image grid, and a graph neural network (GNN) to learn the statistical distribution of physics-guided spatial topological dependency on the contour tree. The two models are co-trained via variational EM. Evaluations on the real-world flood mapping datasets show that the proposed models outperform baseline methods in classification accuracy, especially when training labels are limited. 
    more » « less
  4. The commonsense natural language inference (CNLI) tasks aim to select the most likely follow-up statement to a contextual description of ordinary, everyday events and facts. Current approaches to transfer learning of CNLI models across tasks require many labeled data from the new task. This paper presents a way to reduce this need for additional annotated training data from the new task by leveraging symbolic knowledge bases, such as ConceptNet. We formulate a teacher-student framework for mixed symbolic-neural reasoning, with the large-scale symbolic knowledge base serving as the teacher and a trained CNLI model as the student. This hybrid distillation process involves two steps. The first step is a symbolic reasoning process. Given a collection of unlabeled data, we use an abductive reasoning framework based on Grenander's pattern theory to create weakly labeled data. Pattern theory is an energy-based graphical probabilistic framework for reasoning among random variables with varying dependency structures. In the second step, the weakly labeled data, along with a fraction of the labeled data, is used to transfer-learn the CNLI model into the new task. The goal is to reduce the fraction of labeled data required. We demonstrate the efficacy of our approach by using three publicly available datasets (OpenBookQA, SWAG, and HellaSWAG) and evaluating three CNLI models (BERT, LSTM, and ESIM) that represent different tasks. We show that, on average, we achieve 63% of the top performance of a fully supervised BERT model with no labeled data. With only 1000 labeled samples, we can improve this performance to 72%. Interestingly, without training, the teacher mechanism itself has significant inference power. The pattern theory framework achieves 32.7% accuracy on OpenBookQA, outperforming transformer-based models such as GPT (26.6%), GPT-2 (30.2%), and BERT (27.1%) by a significant margin. We demonstrate that the framework can be generalized to successfully train neural CNLI models using knowledge distillation under unsupervised and semi-supervised learning settings. Our results show that it outperforms all unsupervised and weakly supervised baselines and some early supervised approaches, while offering competitive performance with fully supervised baselines. Additionally, we show that the abductive learning framework can be adapted for other downstream tasks, such as unsupervised semantic textual similarity, unsupervised sentiment classification, and zero-shot text classification, without significant modification to the framework. Finally, user studies show that the generated interpretations enhance its explainability by providing key insights into its reasoning mechanism. 
    more » « less
  5. Batch Normalization (BN) is essential to effectively train state-of-the-art deep Convolutional Neural Networks (CNN). It normalizes the layer outputs during training using the statistics of each mini-batch. BN accelerates training procedure by allowing to safely utilize large learning rates and alleviates the need for careful initialization of the parameters. In this work, we study BN from the viewpoint of Fisher kernels that arise from generative probability models. We show that assuming samples within a mini-batch are from the same probability density function, then BN is identical to the Fisher vector of a Gaussian distribution. That means batch normalizing transform can be explained in terms of kernels that naturally emerge from the probability density function that models the generative process of the underlying data distribution. Consequently, it promises higher discrimination power for the batch-normalized mini-batch. However, given the rectifying non-linearities employed in CNN architectures, distribution of the layer outputs show an asymmetric characteristic. Therefore, in order for BN to fully benefit from the aforementioned properties, we propose approximating underlying data distribution not with one, but a mixture of Gaussian densities. Deriving Fisher vector for a Gaussian Mixture Model (GMM), reveals that batch normalization can be improved by independently normalizing with respect to the statistics of disentangled sub-populations. We refer to our proposed soft piecewise version of batch normalization as Mixture Normalization (MN). Through extensive set of experiments on CIFAR-10 and CIFAR-100, using both a 5-layers deep CNN and modern Inception-V3 architecture, we show that mixture normalization reduces required number of gradient updates to reach the maximum test accuracy of the batch normalized model by ∼31%-47% across a variety of training scenarios. Replacing even a few BN modules with MN in the 48-layers deep Inception-V3 architecture is sufficient to not only obtain considerable training acceleration but also better final test accuracy. We show that similar observations are valid for 40 and 100-layers deep DenseNet architectures as well. We complement our study by evaluating the application of mixture normalization to the Generative Adversarial Networks (GANs), where "mode collapse" hinders the training process. We solely replace a few batch normalization layers in the generator with our proposed mixture normalization. Our experiments using Deep Convolutional GAN (DCGAN) on CIFAR-10 show that mixture normalized DCGAN not only provides an acceleration of ∼58% but also reaches lower (better) "Fréchet Inception Distance" (FID) of 33.35 compared to 37.56 of its batch normalized counterpart. 
    more » « less