Italy

Coded aperture snapshot spectral imaging (CASSI) captures a full frame spectral image as a single compressive image and is mandatory to reconstruct the underlying hyperspectral image (HSI) from the snapshot as the post-processing, which is a challenge inverse problem due to its ill-posed nature. Existing methods for HSI reconstruction from a snapshot usually employs optimization for solving the formulated image degradation model regularized with the empirically designed priors, and still cannot achieve enough reconstruction accuracy for real HS image analysis systems. Motivated by the recent advances of deep learning for different inverse problems, deep learning based HSI reconstruction method has attracted a lot of attention and can boost the reconstruction performance. This study proposes a novel deep convolutional neural network (DCNN) based framework for effectively learning the spatial structure and spectral attribute in the underlying HSI with the reciprocal spatial and spectral modules. Further, to adaptively leverage the useful learned feature for better HSI image reconstruction, we integrate residual attention modules into our DCNN via exploring both spatial and spectral attention maps. Experimental results on two benchmark HSI datasets show that our method outperforms state-of-the-art methods in both quantitative values and visual effects.

ICPR 2020

Deep Residual Attention Network for Hyperspectral Image Reconstruction

hyperspectral image reconstruction deep learning attention

ICPR 2020 is the premier international conference on Pattern Recognition, gathering leading researchers from around the world to present their latest achievements in Pattern Analysis and Recognition, applied Artificial Intelligence and Machine Learning. The topic of the year is " Putting Artificial intelligence to work on Patterns". 
For the first time, ICPR has employed a two-round review process similar to journal submissions, with new papers allowed to be submitted in either the first or second round and papers submitted in the first round and not accepted allowed to be revised and re-submitted for second round review. 

In this edition of the **ICPR 2020** Conference, there are 1407 Oral/Poster presentations organized in five Tracks: 
- Artificial Intelligence, Machine Learning for Pattern Analysis; 
- Biometrics, Human Analysis and Behavior Understanding; Computer Vision, Robotics and Intelligent Systems; 
- Document and Media Analysis; and 
- Image and Signal Processing. 

The **ICPR 2020** Main Conference program also includes 7 Keynotes from prominent world-renowned scientists, 8 Tutorials, 28 Demos, 8 Challenges, 2 Panels, as well as an Industry/Sponsor Exhibition. In addition, 41 Workshops on a range of important current and emerging research topics complete the ICPR 2020 program. 


poster

Segmentation of kidney on CT images is critical to computer-assisted surgical planning for kidney interventional therapy. Segmenting kidney manually is impractical in clinical, automatic segmentation is desirable. U-Net has been successful in medical image segmentation and is a promising candidate for the task. However, semantic gap still exists, especially when multiple phase images or multiple center images are involved. In this paper, we proposed an ULBNet to reduce the semantic gap and to improve segmentation performance. The proposed architecture includes new skip connections of local binary convolution (LBC). We also proposed a novel strategy of fast retraining a model for a new task without manually labelling required. We evaluated the network for kidney segmentation on multiple phase CT images. ULBNet resulted in an overall accuracy of 98.0% with comparison to Resunet 97.5%. Specifically, on the plain phase CT images, 98.1% resulted from ULBNet and 97.6% from Resunet; on the corticomedullay phase images, 97.8% from ULBNet and 97.2% from Resunet; on the nephrographic phase images, 97.6% from ULBNet and 97.4% from Resunet; on the excretory phase images, 98.1% from ULBNet and 97.4% from Resunet. The proposed network architecture performs better than Resunet on generalizing to multiple phase images.

Segmenting Kidney on Multiple Phase CT Images using ULBNet

The performance of deep learning models are driven by various parameters but to tune all of them every time, for every given dataset, is a heuristic practice. In this paper, unlike the common practice of decaying the learning rate, we propose a step-wise training strategy where the learning rate and the batch size are tuned based on the dataset size. Here, the given dataset size is progressively increased during the training to boost the network performance without saturating the learning curve, which is seen after certain epochs. We conducted extensive experiments on multiple networks and datasets to validate the proposed training strategy. The experimental results proves our hypothesis that the learning rate, the batch size and the data size are interrelated and can improve the network accuracy if an optimal progressive step-wise training strategy is applied. The proposed strategy also reduces the overall training cost compared to the baseline approach.

Rethinking of Deep Models Parameters with Respect to Data Distribution

Current techniques to analyze eye-tracking data for diagnosing neuropsychiatric disorders with visual scanning tests are largely non-generalizable and rely heavily on domain expertise and study-specific assumptions. We present a general, data-driven, end-to-end framework, which when tested on two neuropsychiatric study datasets, exceed the performance of state-of-the-art, study-specific analysis models.

A General End-to-End Method for Characterizing Neuropsychiatric Disorders using Free-Viewing Visual Scanning Tasks

Action recognition in still images is a challenging task because of the complexity of human motions and the variance of background in the same action category. And some actions typically occur in fine-grained categories, with little visual differences between these categories. So extracting discriminative features or modeling various semantic parts is essential for image-based action recognition. Many methods apply expensive manual annotations to learn discriminative parts information for action recognition, which may severely hinder potential applications in real life. In recent years, bilinear pooling method has shown its effectiveness for image classification due to its learning distinctive features automatically. Inspired by this model, in this paper, an improved bilinear pooling method is proposed for image-based action recognition. The previous bilinear pooling approaches contain lots of noisy background or harmful feature information, which limit their application for action recognition. In our method, 
the attention mechanism is introduced into hierarchical bilinear pooling framework with mask aggregation. The proposed model can generate the distinctive and RoI-aware feature information by combining multiple attention mask maps from the channel and spatial-wise attention features. Specifically, our method makes the network to pay more attention to discriminative region of the vital objects in an image. We verify our model on the two challenging datasets: 1) Stanford 40 action dataset and 2) our action dataset that includes 60 categories. Experimental results demonstrate the effectiveness of our approach, which is superior to the traditional and state-of-the-art methods.

An Improved Bilinear Pooling Method for Image- Based Action Recognition

We revisit the long-standing problem of learning
from a small sample, to which end we propose a novel method called
GLICO (Generative Latent Implicit Conditional Optimization).
GLICO learns a mapping from the training examples to a latent
space, and a generator that generates images from vectors in
the latent space. Unlike most recent works, which rely on access to
large amounts of unlabeled data, GLICO does not require access
to any additional data other than the small set of labeled points.
In fact, GLICO learns to synthesize completely new samples for
every class using as little as 5 or 10 examples per class, with as few
as 10 such classes without imposing any prior. GLICO is then used
to augment the small training set while training a classifier on
the small sample. To this end, our proposed method samples the
learned latent space using spherical interpolation and generates
new examples using the trained generator. Empirical results
show that the new sampled set is diverse enough, leading to
improvement in image classification in comparison with the state
of the art, when trained on small samples obtained from CIFAR10, CIFAR-100, and CUB-200.

Generative Latent Implicit Conditional Optimization when Learning from Small Sample


Deep neural networks have achieved remarkable results on image super-resolution (SR), but the efficiency problem of deep SR networks is rarely studied. We experimentally find that many sequentially stacked convolutional blocks in nowadays SR networks are far from being fully optimized, which largely damages their overall efficiency. It indicates that comparable or even better results could be achieved with less but sufficiently optimized blocks. In this paper, we try to construct a more efficient SR model via the proposed recursive aggregation network (RAN). It recursively aggregates convolutional blocks in different orders and avoids too many sequentially stacked blocks. In this way, multiple shortcuts are introduced in RAN, and help gradients easier flow to all inner layers, even for very deep SR networks. As a result, all blocks in RAN can be better optimized, thus RAN can achieve better performance with a smaller model size than existing methods

Efficient Super-Resolution by Recursive Aggregation

Work in the construction site is considered to be one of the occupations with the highest safety risk factor. Therefore, safety plays an important role in construction site. One of the most fundamental safety rules in construction site is to wear a hardhat. To strengthen the safety of the construction site, most of the current methods use multi-stage method for hardhat-wearing detection. These methods have limitations in terms of adaptability and generalizability. In this paper, we propose a one-stage object detection method based on convolutional neural network. We present a multi-scale strategy that selects the high-resolution feature maps of DarkNet-53 to effectively identify small-scale hardhats. In addition, we propose an improved weighted bi-directional feature pyramid network (BiFPN), which could fuse more semantic features from more scales. The proposed method can not only detect hardhat-wearing, but also identify the color of the hardhat. Experimental results show that the proposed method achieves a mAP of 87.04%, which outperforms several state-of-the-art methods on a public dataset.

Construction worker hardhat-wearing detection based on an improved BiFPN

The adaptation of deep convolutional neural network has made revolutionary advances in human body posture estimation. Various applications utilizing deep neural network for pose estimation have drawn considerable attention in recent years. However, prediction and localization of keypoints in single-person and multi-person images is still a challenging problem. Towards this, we propose a bottom-up approach to pose estimation and motion recognition. We present StrongPose system that deals with object-part associations using part-based modeling. The convolution network in our model detects strong keypoint heat maps and predicts their comparative displacements, allowing keypoints to be grouped into human instances. Further, it utilizes the keypoints to generate body heat maps that can determine the position of the human body in the image. The StrongPose system is based on fully convolutional engineering and makes proficient inferences while maintaining runtime regardless of the number of individuals in the image. We train and test the StrongPose on the COCO dataset. Evaluation results show that our framework achieves average precision of 0.708 using ResNet-101 and 0.725 using ResNet-152. Our results considerably outperform prior bottom-up frameworks.

StrongPose: Bottom-up and Strong Keypoint Heat Map Based Pose Estimation

Video moment retrieval aims to search the moment most relevant to a given language query. However, most existing methods in this community often require temporal boundary annotations which are expensive and time-consuming to label. Hence weakly supervised methods have been put forward recently by only using coarse video-level label. Despite effectiveness, these methods usually process moment candidates independently, while ignoring a critical issue that the natural temporal dependencies between candidates in different temporal scales. To cope with this issue, we propose a Multi-scale 2D Representation Learning method for weakly supervised video moment retrieval. Specifically, we first construct a two-dimensional map for each temporal scale to capture the temporal dependencies between candidates. Two dimensions in this map indicate the start and end time points of these candidates. Then, we select top-K candidates from each scale-varied map with a learnable convolutional neural network. With a newly designed Moments Evaluation Module, we obtain the alignment scores of the selected candidates. At last, the similarity between captions and language query is served as supervision for further training the candidates' selector. Experiments on two benchmark datasets Charades-STA and ActivityNet Captions demonstrate that our approach achieves superior performance to state-of-the-art results.

Multi-scale 2D Representation Learning for weakly-supervised moment retrieval


Face age editing has become a crucial task in film post-production, and is also becoming popular for general purpose photography. Recently, adversarial training has produced some of the most visually impressive results for image manipulation, including the face aging/de-aging task. In spite of considerable progress, current methods often present visual artifacts and can only deal with low-resolution images. In order to achieve aging/de-aging with the high quality and robustness necessary for wider use, these problems need to be addressed. This is the goal of the present work. We present an encoder-decoder architecture for face age editing. The core idea of our network is to encode a face image to age-invariant features, and learn a modulation vector corresponding to a target age. We then combine these two elements to produce a realistic image of the person with the desired target age. Our architecture is greatly simplified with respect to other approaches, and allows for fine-grained age editing on high resolution images in a single unified model. Source codes are available at https://github.com/InterDigitalInc/HRFAE.

Premium content

Downloads

Next from ICPR 2020

Segmenting Kidney on Multiple Phase CT Images using ULBNet

Similar lecture

Snapshot Hyperspectral Imaging Based on Weighted High-order Singular Value Regularization