arXiv subjects

Jan van Gemert

Publications and source records attributed to Jan van Gemert.

At least 73 records · Page 4Linked to original sources

arXiv2020

Real-time Webcam Heart-Rate and Variability Estimation with Clean Ground Truth for Evaluation

Remote photo-plethysmography (rPPG) uses a camera to estimate a person's heart rate (HR). Similar to how heart rate can provide useful information about a person's vital signs, insights about the underlying physio/psychological conditions can be obtained from heart rate variability (HRV). HRV is a measure of the fine fluctuations in the intervals between heart beats. However, this measure requires temporally locating heart beats with a high degree of precision. We introduce a refined and efficient real-time rPPG pipeline with novel filtering and motion suppression that not only estimates heart rates, but also extracts the pulse waveform to time heart beats and measure heart rate variability. This unsupervised method requires no rPPG specific training and is able to operate in real-time. We also introduce a new multi-modal video dataset, VicarPPG 2, specifically designed to evaluate rPPG algorithms on HR and HRV estimation. We validate and study our method under various conditions on a comprehensive range of public and self-recorded datasets, showing state-of-the-art results and providing useful insights into some unique aspects. Lastly, we make available CleanerPPG, a collection of human-verified ground truth peak/heart-beat annotations for existing rPPG datasets. These verified annotations should make future evaluations and benchmarking of rPPG algorithms more accurate, standardized and fair.

cs.CV↗

arXiv2020

Deep Unsupervised Image Hashing by Maximizing Bit Entropy

Unsupervised hashing is important for indexing huge image or video collections without having expensive annotations available. Hashing aims to learn short binary codes for compact storage and efficient semantic retrieval. We propose an unsupervised deep hashing layer called Bi-half Net that maximizes entropy of the binary codes. Entropy is maximal when both possible values of the bit are uniformly (half-half) distributed. To maximize bit entropy, we do not add a term to the loss function as this is difficult to optimize and tune. Instead, we design a new parameter-free network layer to explicitly force continuous image features to approximate the optimal half-half bit distribution. This layer is shown to minimize a penalized term of the Wasserstein distance between the learned continuous image features and the optimal half-half bit distribution. Experimental results on the image datasets Flickr25k, Nus-wide, Cifar-10, Mscoco, Mnist and the video datasets Ucf-101 and Hmdb-51 show that our approach leads to compact codes and compares favorably to the current state-of-the-art.

cs.CV↗

arXiv2020

ReproducedPapers.org: Openly teaching and structuring machine learning reproducibility

We present ReproducedPapers.org: an open online repository for teaching and structuring machine learning reproducibility. We evaluate doing a reproduction project among students and the added value of an online reproduction repository among AI researchers. We use anonymous self-assessment surveys and obtained 144 responses. Results suggest that students who do a reproduction project place more value on scientific reproductions and become more critical thinkers. Students and AI researchers agree that our online reproduction repository is valuable.

cs.CY↗

arXiv2020

Zoom-CAM: Generating Fine-grained Pixel Annotations from Image Labels

Current weakly supervised object localization and segmentation rely on class-discriminative visualization techniques to generate pseudo-labels for pixel-level training. Such visualization methods, including class activation mapping (CAM) and Grad-CAM, use only the deepest, lowest resolution convolutional layer, missing all information in intermediate layers. We propose Zoom-CAM: going beyond the last lowest resolution layer by integrating the importance maps over all activations in intermediate layers. Zoom-CAM captures fine-grained small-scale objects for various discriminative class instances, which are commonly missed by the baseline visualization methods. We focus on generating pixel-level pseudo-labels from class labels. The quality of our pseudo-labels evaluated on the ImageNet localization task exhibits more than 2.8% improvement on top-1 error. For weakly supervised semantic segmentation our generated pseudo-labels improve a state of the art model by 1.1%.

cs.CV↗

arXiv2020

Respecting Domain Relations: Hypothesis Invariance for Domain Generalization

In domain generalization, multiple labeled non-independent and non-identically distributed source domains are available during training while neither the data nor the labels of target domains are. Currently, learning so-called domain invariant representations (DIRs) is the prevalent approach to domain generalization. In this work, we define DIRs employed by existing works in probabilistic terms and show that by learning DIRs, overly strict requirements are imposed concerning the invariance. Particularly, DIRs aim to perfectly align representations of different domains, i.e. their input distributions. This is, however, not necessary for good generalization to a target domain and may even dispose of valuable classification information. We propose to learn so-called hypothesis invariant representations (HIRs), which relax the invariance assumptions by merely aligning posteriors, instead of aligning representations. We report experimental results on public domain generalization datasets to show that learning HIRs is more effective than learning DIRs. In fact, our approach can even compete with approaches using prior knowledge about domains.

cs.LG↗

arXiv2020

WeightAlign: Normalizing Activations by Weight Alignment

Batch normalization (BN) allows training very deep networks by normalizing activations by mini-batch sample statistics which renders BN unstable for small batch sizes. Current small-batch solutions such as Instance Norm, Layer Norm, and Group Norm use channel statistics which can be computed even for a single sample. Such methods are less stable than BN as they critically depend on the statistics of a single input sample. To address this problem, we propose a normalization of activation without sample statistics. We present WeightAlign: a method that normalizes the weights by the mean and scaled standard derivation computed within a filter, which normalizes activations without computing any sample statistics. Our proposed method is independent of batch size and stable over a wide range of batch sizes. Because weight statistics are orthogonal to sample statistics, we can directly combine WeightAlign with any method for activation normalization. We experimentally demonstrate these benefits for classification on CIFAR-10, CIFAR-100, ImageNet, for semantic segmentation on PASCAL VOC 2012 and for domain adaptation on Office-31.

cs.CV↗

arXiv2020

Efficiency in Real-time Webcam Gaze Tracking

Efficiency and ease of use are essential for practical applications of camera based eye/gaze-tracking. Gaze tracking involves estimating where a person is looking on a screen based on face images from a computer-facing camera. In this paper we investigate two complementary forms of efficiency in gaze tracking: 1. The computational efficiency of the system which is dominated by the inference speed of a CNN predicting gaze-vectors; 2. The usability efficiency which is determined by the tediousness of the mandatory calibration of the gaze-vector to a computer screen. To do so, we evaluate the computational speed/accuracy trade-off for the CNN and the calibration effort/accuracy trade-off for screen calibration. For the CNN, we evaluate the full face, two-eyes, and single eye input. For screen calibration, we measure the number of calibration points needed and evaluate three types of calibration: 1. pure geometry, 2. pure machine learning, and 3. hybrid geometric regression. Results suggest that a single eye input and geometric regression calibration achieve the best trade-off.

cs.CV↗

arXiv2020

Black Magic in Deep Learning: How Human Skill Impacts Network Training

How does a user's prior experience with deep learning impact accuracy? We present an initial study based on 31 participants with different levels of experience. Their task is to perform hyperparameter optimization for a given deep learning architecture. The results show a strong positive correlation between the participant's experience and the final performance. They additionally indicate that an experienced participant finds better solutions using fewer resources on average. The data suggests furthermore that participants with no prior experience follow random strategies in their pursuit of optimal hyperparameters. Our study investigates the subjective human factor in comparisons of state of the art results and scientific reproducibility in deep learning.

cs.CV↗

arXiv2020

Evaluating the performance of the LIME and Grad-CAM explanation methods on a LEGO multi-label image classification task

In this paper, we run two methods of explanation, namely LIME and Grad-CAM, on a convolutional neural network trained to label images with the LEGO bricks that are visible in them. We evaluate them on two criteria, the improvement of the network's core performance and the trust they are able to generate for users of the system. We find that in general, Grad-CAM seems to outperform LIME on this specific task: it yields more detailed insight from the point of view of core performance and 80\% of respondents asked to choose between them when it comes to the trust they inspire in the model choose Grad-CAM. However, we also posit that it is more useful to employ these two methods together, as the insights they yield are complementary.

cs.CV↗

arXiv2020

Rethinking Online Action Detection in Untrimmed Videos: A Novel Online Evaluation Protocol

The Online Action Detection (OAD) problem needs to be revisited. Unlike traditional offline action detection approaches, where the evaluation metrics are clear and well established, in the OAD setting we find very few works and no consensus on the evaluation protocols to be used. In this work we propose to rethink the OAD scenario, clearly defining the problem itself and the main characteristics that the models which are considered online must comply with. We also introduce a novel metric: the Instantaneous Accuracy ($IA$). This new metric exhibits an \emph{online} nature and solves most of the limitations of the previous metrics. We conduct a thorough experimental evaluation on 3 challenging datasets, where the performance of various baseline methods is compared to that of the state-of-the-art. Our results confirm the problems of the previous evaluation protocols, and suggest that an IA-based protocol is more adequate to the online scenario. The baselines models and a development kit with the novel evaluation protocol are publicly available: https://github.com/gramuah/ia.

cs.CV↗

arXiv2020

The Instantaneous Accuracy: a Novel Metric for the Problem of Online Human Behaviour Recognition in Untrimmed Videos

The problem of Online Human Behaviour Recognition in untrimmed videos, aka Online Action Detection (OAD), needs to be revisited. Unlike traditional offline action detection approaches, where the evaluation metrics are clear and well established, in the OAD setting we find few works and no consensus on the evaluation protocols to be used. In this paper we introduce a novel online metric, the Instantaneous Accuracy ($IA$), that exhibits an \emph{online} nature, solving most of the limitations of the previous (offline) metrics. We conduct a thorough experimental evaluation on TVSeries dataset, comparing the performance of various baseline methods to the state of the art. Our results confirm the problems of previous evaluation protocols, and suggest that an IA-based protocol is more adequate to the online scenario for human behaviour understanding. Code of the metric available https://github.com/gramuah/ia

cs.CV↗

arXiv2019

Attention-Aware Age-Agnostic Visual Place Recognition

A cross-domain visual place recognition (VPR) task is proposed in this work, i.e., matching images of the same architectures depicted in different domains. VPR is commonly treated as an image retrieval task, where a query image from an unknown location is matched with relevant instances from geo-tagged gallery database. Different from conventional VPR settings where the query images and gallery images come from the same domain, we propose a more common but challenging setup where the query images are collected under a new unseen condition. The two domains involved in this work are contemporary street view images of Amsterdam from the Mapillary dataset (source domain) and historical images of the same city from Beeldbank dataset (target domain). We tailored an age-invariant feature learning CNN that can focus on domain invariant objects and learn to match images based on a weakly supervised ranking loss. We propose an attention aggregation module that is robust to domain discrepancy between the train and the test data. Further, a multi-kernel maximum mean discrepancy (MK-MMD) domain adaptation loss is adopted to improve the cross-domain ranking performance. Both attention and adaptation modules are unsupervised while the ranking loss uses weak supervision. Visual inspection shows that the attention module focuses on built forms while the dramatically changing environment are less weighed. Our proposed CNN achieves state of the art results (99% accuracy) on the single-domain VPR task and 20% accuracy at its best on the cross-domain VPR task, revealing the difficulty of age-invariant VPR.

cs.CV↗

arXiv2019

Running Event Visualization using Videos from Multiple Cameras

Visualizing the trajectory of multiple runners with videos collected at different points in a race could be useful for sports performance analysis. The videos and the trajectories can also aid in athlete health monitoring. While the runners unique ID and their appearance are distinct, the task is not straightforward because the video data does not contain explicit information as to which runners appear in each of the videos. There is no direct supervision of the model in tracking athletes, only filtering steps to remove irrelevant detections. Other factors of concern include occlusion of runners and harsh illumination. To this end, we identify two methods for runner identification at different points of the event, for determining their trajectory. One is scene text detection which recognizes the runners by detecting a unique 'bib number' attached to their clothes and the other is person re-identification which detects the runners based on their appearance. We train our method without ground truth but to evaluate the proposed methods, we create a ground truth database which consists of video and frame interval information where the runners appear. The videos in the dataset was recorded by nine cameras at different locations during the a marathon event. This data is annotated with bib numbers of runners appearing in each video. The bib numbers of runners known to occur in the frame are used to filter irrelevant text and numbers detected. Except for this filtering step, no supervisory signal is used. The experimental evidence shows that the scene text recognition method achieves an F1-score of 74. Combining the two methods, that is - using samples collected by text spotter to train the re-identification model yields a higher F1-score of 85.8. Re-training the person re-identification model with identified inliers yields a slight improvement in performance(F1 score of 87.8).

cs.CV↗

arXiv2019

Efficient Real-Time Camera Based Estimation of Heart Rate and Its Variability

Remote photo-plethysmography (rPPG) uses a remotely placed camera to estimating a person's heart rate (HR). Similar to how heart rate can provide useful information about a person's vital signs, insights about the underlying physio/psychological conditions can be obtained from heart rate variability (HRV). HRV is a measure of the fine fluctuations in the intervals between heart beats. However, this measure requires temporally locating heart beats with a high degree of precision. We introduce a refined and efficient real-time rPPG pipeline with novel filtering and motion suppression that not only estimates heart rate more accurately, but also extracts the pulse waveform to time heart beats and measure heart rate variability. This method requires no rPPG specific training and is able to operate in real-time. We validate our method on a self-recorded dataset under an idealized lab setting, and show state-of-the-art results on two public dataset with realistic conditions (VicarPPG and PURE).

cs.CV↗

arXiv2019

Push for Quantization: Deep Fisher Hashing

Current massive datasets demand light-weight access for analysis. Discrete hashing methods are thus beneficial because they map high-dimensional data to compact binary codes that are efficient to store and process, while preserving semantic similarity. To optimize powerful deep learning methods for image hashing, gradient-based methods are required. Binary codes, however, are discrete and thus have no continuous derivatives. Relaxing the problem by solving it in a continuous space and then quantizing the solution is not guaranteed to yield separable binary codes. The quantization needs to be included in the optimization. In this paper we push for quantization: We optimize maximum class separability in the binary space. We introduce a margin on distances between dissimilar image pairs as measured in the binary space. In addition to pair-wise distances, we draw inspiration from Fisher's Linear Discriminant Analysis (Fisher LDA) to maximize the binary distances between classes and at the same time minimize the binary distance of images within the same class. Experiments on CIFAR-10, NUS-WIDE and ImageNet100 demonstrate compact codes comparing favorably to the current state of the art.

cs.CV↗

arXiv2019

Deep Visual City Recognition Visualization

Understanding how cities visually differ from each others is interesting for planners, residents, and historians. We investigate the interpretation of deep features learned by convolutional neural networks (CNNs) for city recognition. Given a trained city recognition network, we first generate weighted masks using the known Grad-CAM technique and to select the most discriminate regions in the image. Since the image classification label is the city name, it contains no information of objects that are class-discriminate, we investigate the interpretability of deep representations with two methods. (i) Unsupervised method is used to cluster the objects appearing in the visual explanations. (ii) A pretrained semantic segmentation model is used to label objects in pixel level, and then we introduce statistical measures to quantitatively evaluate the interpretability of discriminate objects. The influence of network architectures and random initializations in training, is studied on the interpretability of CNN features for city recognition. The results suggest that network architectures would affect the interpretability of learned visual representations greater than different initializations.

cs.CV↗

arXiv2019

ViDeNN: Deep Blind Video Denoising

We propose ViDeNN: a CNN for Video Denoising without prior knowledge on the noise distribution (blind denoising). The CNN architecture uses a combination of spatial and temporal filtering, learning to spatially denoise the frames first and at the same time how to combine their temporal information, handling objects motion, brightness changes, low-light conditions and temporal inconsistencies. We demonstrate the importance of the data used for CNNs training, creating for this purpose a specific dataset for low-light conditions. We test ViDeNN on common benchmarks and on self-collected data, achieving good results comparable with the state-of-the-art.

cs.CV↗

arXiv2017

Object-Extent Pooling for Weakly Supervised Single-Shot Localization

In the face of scarcity in detailed training annotations, the ability to perform object localization tasks in real-time with weak-supervision is very valuable. However, the computational cost of generating and evaluating region proposals is heavy. We adapt the concept of Class Activation Maps (CAM) into the very first weakly-supervised 'single-shot' detector that does not require the use of region proposals. To facilitate this, we propose a novel global pooling technique called Spatial Pyramid Averaged Max (SPAM) pooling for training this CAM-based network for object extent localisation with only weak image-level supervision. We show this global pooling layer possesses a near ideal flow of gradients for extent localization, that offers a good trade-off between the extremes of max and average pooling. Our approach only requires a single network pass and uses a fast-backprojection technique, completely omitting any region proposal steps. To the best of our knowledge, this is the first approach to do so. Due to this, we are able to perform inference in real-time at 35fps, which is an order of magnitude faster than all previous weakly supervised object localization frameworks.

cs.CV↗