Italy

Wearable cameras are becoming more and more popular in several applications, increasing the interest of the research community in developing approaches for recognizing actions from the first-person point of view. An open challenge in egocentric action recognition is that videos lack detailed information about the main actor&#39;s pose and thus tend to record only parts of the movement when focusing on manipulation tasks. Thus, the amount of information about the action itself is limited, making crucial the understanding of the manipulated objects and their context. Many previous works addressed this issue with two-stream architectures, where one stream is dedicated to modeling the appearance of objects involved in the action, and another to extracting motion features from optical flow. In this paper, we argue that learning features jointly from these two information channels is beneficial to capture the spatio-temporal correlations between the two better. To this end, we propose a single stream architecture able to do so, thanks to the addition of a self-supervised block that uses a pretext motion prediction task to intertwine motion and appearance knowledge. Experiments on several publicly available databases show the power of our approach.

ICPR 2020

Self-Supervised Joint Encoding of Motion and Appearance for First Person Action Recognition

first-person action recognition

motion segmentation

self-supervision

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. 


Wearable cameras are becoming more and more popular in several applications, increasing the interest of the research community in developing approaches for recognizing actions from the first-person point of view. An open challenge in egocentric action recognition is that videos lack detailed information about the main actor's pose and thus tend to record only parts of the movement when focusing on manipulation tasks. Thus, the amount of information about the action itself is limited, making crucial the understanding of the manipulated objects and their context. Many previous works addressed this issue with two-stream architectures, where one stream is dedicated to modeling the appearance of objects involved in the action, and another to extracting motion features from optical flow. In this paper, we argue that learning features jointly from these two information channels is beneficial to capture the spatio-temporal correlations between the two better. To this end, we propose a single stream architecture able to do so, thanks to the addition of a self-supervised block that uses a pretext motion prediction task to intertwine motion and appearance knowledge. Experiments on several publicly available databases show the power of our approach.

poster


This paper studies the problem of Person Re-Identification (ReID)
for large-scale applications. Recent research efforts have been devoted
to building complicated part models, which introduce considerably
high computational cost and memory consumption, inhibiting its practicability
in large-scale applications. This paper aims to develop a novel learning
strategy to find efficient feature embeddings while maintaining the
balance of accuracy and model complexity. More specifically, we find
by enhancing the classical triplet loss together with cross-entropy
loss, our method can explore the hard examples and build a discriminant
feature embedding yet compact enough for large-scale applications.
Our method is carried out progressively using Bayesian optimization,
and we call it the Progressive Learning Algorithm (PLA). Extensive
experiments on three large-scale datasets show that our PLA is comparable
or better than the state-of-the-arts. Especially, on the challenging
Market-1501 dataset, we achieve Rank-1=94.7%/mAP=89.4% while saving
at least 30\% parameters than strong part models.

Probabilistic Word Embeddings in Kinematic Space


In this paper, an end-to-end neural embedding system based on triplet loss and residual learning has been proposed for speech emotion recognition. The proposed system learns the embeddings from the emotional information of the speech utterances. The learned embeddings are used to recognize the emotions portrayed by given speech samples of various lengths. The proposed system implements Residual Neural Network architecture. It is trained using softmax pre-training and triplet loss function. The weights between the fully connected and embedding layers of the trained network are used to calculate the embedding values. The embedding representations of various emotions are mapped onto a hyperplane, and the angles among them are computed using the cosine similarity. These angles are utilized to classify a new speech sample into its appropriate emotion class. The proposed system has demonstrated 91.67% and 64.44% accuracy while recognizing emotions for RAVDESS and IEMOCAP dataset, respectively.

End-to-end Triplet Loss based Emotion Embedding System for Speech Emotion Recognition


Lung segmentation on CT images is a crucial step for a computer-aided diagnosis system of lung diseases. The existing deep learning based lung segmentation methods are less efficient to segment lungs on clinical CT images, especially that the segmentation on lung boundaries is not accurate enough due to complex pulmonary opacities in practical clinics. In this paper, we propose a boundary-guided network (BG-Net) to address this problem. It contains two auxiliary branches that seperately segment lungs and extract the lung boundaries, and an aggregation branch that efficiently exploits lung boundary cues to guide the network for more accurate lung segmentation on clinical CT images. We evaluate the proposed method on a private dataset collected from the Osaka university hospital and four public datasets including StructSeg, HUG, VESSEL12, and a Novel Coronavirus 2019 (COVID-19) dataset. Experimental results show that the proposed method can segment lungs more accurately and outperform several other deep learning based methods.

BG-Net: Boundary-Guided Network for Lung Segmentation on Clinical CT Images


Low-dose CT imaging emerges with low radiation risk due to the reduction of radiation dose, but brings negative impact on the imaging quality. This paper addresses the problem of low-dose CT reconstruction. Previous methods are
unsatisfactory due to the inaccurate recovery of image details under the strong noise generated by the reduction of radiation dose, which directly affects the final diagnosis. To suppress the noise effectively while retain the structures well, we propose a detail-revealing dual-branch aggregation network to effectively reconstruct the degraded CT image. Specifically, the main reconstruction branch iteratively exploits and compensates the reconstruction errors to gradually refine the CT image, while the prior branch is to learn the structure details as prior knowledge to help recover the CT image. A sophisticated detail-revealing loss is designed to fuse the information from both branches and guide the learning to obtain better performance from pixel-wise and holistic perspectives respectively. Experimental results show that our method outperforms the state-of-art methods in both PSNR and SSIM metrics.

Detail-Revealing Deep Low-Dose CT Reconstruction

Aspect-Based Sentiment Analysis (ABSA) studies the extraction of sentiments and their targets. Collecting labeled data for this task in order to help neural networks generalize better can be laborious and time-consuming. As an alternative, similar data to the real-world examples can be produced artificially through an adversarial process which is carried out in the embedding space. Although these examples are not real sentences, they have been shown to act as a regularization method which can make neural networks more robust. In this work, we fine-tune the general purpose BERT and domain specific post-trained BERT (BERT-PT) using adversarial training. After improving the results of post-trained BERT with different hyperparameters, we propose a novel architecture called BERT Adversarial Training (BAT) to utilize adversarial training for the two major tasks of Aspect Extraction and Aspect Sentiment Classification in sentiment analysis. The proposed model outperforms the general BERT as well as the in-domain post-trained BERT in both tasks. To the best of our knowledge, this is the first study on the application of adversarial training in ABSA.

Adversarial Training for Aspect-Based Sentiment Analysis with BERT

In this paper we propose a method for improving
pedestrian detection in the thermal domain using two stages:
first, a generative data augmentation approach is used, then
a domain adaptation method using generated data adapts an
RGB pedestrian detector. Our model, based on the Least-Squares
Generative Adversarial Network, is trained to synthesize realistic
thermal versions of input RGB images which are then used to
augment the limited amount of labeled thermal pedestrian images
available for training. We apply our generative data augmentation strategy in order to adapt a pretrained YOLOv3 pedestrian
detector to detection in the thermal-only domain. Experimental
results demonstrate the effectiveness of our approach: using less
than 50% of available real thermal training data, and relying
on synthesized data generated by our model in the domain
adaptation phase, our detector achieves state-of-the-art results on
the KAIST Multispectral Pedestrian Detection Benchmark; even
if more real thermal data is available adding GAN generated
images to the training data results in improved performance,
thus showing that these images act as an effective form of data
augmentation. To the best of our knowledge, our detector achieves
the best single-modality detection results on KAIST with respect
to the state-of-the-art.

Robust pedestrian detection in thermal imagery using synthesized images

Convolutional neural network training can suffer from diverse issues like exploding or vanishing gradients, scaling-based weight space symmetry and covariant-shift. In order to address these issues, researchers develop weight regularization methods and activation normalization methods. In this work we propose a weight soft-regularization method based on the Oblique manifold. The proposed method uses a loss function which pushes each weight vector to have a norm close to one, i.e. the weight matrix is smoothly steered toward the so-called Oblique manifold. We evaluate our method on the very popular CIFAR-10, CIFAR-100 and ImageNet 2012 datasets using two state-of-the-art architectures, namely the ResNet and wide-ResNet. Our method introduces negligible computational overhead and the results show that it is competitive to the state-of-the-art and in some cases superior to it. Additionally, the results are less sensitive to hyperparameter settings such as batch size and regularization factor.

Norm Loss: An efficient yet effective regularization method for deep neural networks


Video stabilization is fundamental for providing good Quality of Experience for viewers and generating suitable content for video applications. In this scenario, Digital Video Stabilization (DVS) is convenient and economical for casual or amateur recording because it neither requires specific equipment nor demands knowledge of the device used for recording. Although DVS has been a research topic for decades, with a number of proposals from industry and academia, traditional methods tend to fail in a number of scenarios, e.g. with occlusion, textureless areas, parallax, dark areas, amongst others. On the other hand, defining a smooth camera path is a hard task in Deep Learning scenarios. This paper proposes a video stabilization system based on traditional and Deep Learning methods. First, we leverage Spatial Transformer Networks (STNs) to learn transformation parameters between image pairs, then utilize this knowledge to stabilize videos: we obtain the motion parameters between frame pairs and then smooth the camera path using moving averages. Our approach aims at combining the strengths of both Deep Learning and traditional methods: the ability of STNs to estimate motion parameters between two frames and the effectiveness of moving averages to smooth camera paths. Experimental results show that our system outperforms state-of-the-art proposals and a commercial solution.

Deep Homography-Based Video Stabilization​


In this paper, we propose an automatic approach for localizing the inner eye canthus in thermal face images. We first coarsely detect 5 facial keypoints corresponding to the center of the eyes, the nose tip, and the ears. Then we compute a sparse 2D-3D points correspondence using a 3D Morphable Face Model (3DMM). This correspondence is used to project the entire 3D face onto the image, and subsequently locate the inner eye canthus. Detecting this location allows obtaining the most precise body temperature measurement for a person using a thermal camera. We evaluated the approach on a thermal face dataset provided with manually annotated landmarks. However, such manual annotations are normally conceived to identify facial parts such as eyes, nose, and mouth, and are not specifically tailored for localizing the eye canthus region. As an additional contribution, we enrich the original dataset by using the annotated landmarks to deform and project the 3DMM onto the images. Then, by manually selecting a small region corresponding to the eye canthus, we enrich the dataset with additional annotations. By using the manual landmarks, we ensure the correctness of the 3DMM projection, which can be used as ground-truth for future evaluations. Moreover, we supply the dataset with the 3D head poses and per-point visibility masks for detecting self-occlusions.

Inner Eye Canthus Localization for Human Body Temperature Screening


In this paper, we seek to extend multi-object tracking research on a relatively less explored domain, that of, underwater multi-object tracking in the wild. Multi-object fish tracking is an important task because it can provide fish monitoring systems with richer information (e.g. multiple views of the same fish) as compared to detections and it can be an invaluable input to fish behavior analysis. However, there is a lack of an annotated benchmark dataset with enough samples for this task. To circumvent the need for manual ground truth tracking annotation, we craft a synthetic dataset. Using this synthetic dataset, we train an integrated detector and tracker called SynDHN. SynDHN uses the Deep Hungarian Network (DHN), which is a differentiable approximation of the Hungarian assignment algorithm. We repurpose DHN to become the tracking component of our algorithm by performing the task of affinity estimation between detector predictions. We consider both spatial and appearance features for affinity estimation. Our results show that despite being trained on a synthetic dataset, SynDHN generalizes well to real underwater video tracking and performs better against our baseline algorithms. 


Premium content

Self-Supervised Joint Encoding of Motion and Appearance for First Person Action Recognition

Downloads

Next from ICPR 2020

Probabilistic Word Embeddings in Kinematic Space

Similar lecture

Motion supervised co-part segmentation