This paper presents OpenREALM, a real-time mapping framework for Unmanned Aerial Vehicles (UAVs). A camera attached to the onboard computer of a moving UAV is utilized to acquire high resolution image mosaics of a targeted area of interest. Different modes of operation allow OpenREALM to perform simple stitching assuming an approximate plane ground, or to fully recover complex 3D surface information to extract both elevation maps and geometrically corrected orthophotos. Additionally, the global position of the UAV is used to georeference the data. In all modes incremental progress of the resulting map can be viewed live by an operator on the ground. Obtained, up-to-date surface information will be a push forward to a variety of UAV applications. For the benefit of the community, source code is public at https://github.com/laxnpander/OpenREALM.
In this paper, we present an Improved Data Augmentation (IDA) technique focused on Salient Object Detection (SOD). Standard data augmentation techniques proposed in the literature, such as image cropping, rotation, flipping, and resizing, only generate variations of the existing examples, providing a limited generalization. Our method combines image inpainting, affine transformations, and the linear combination of different generated background images with salient objects extracted from labeled data. Our proposed technique enables more precise control of the object's position and size while preserving background information. The background choice is based on an inter-image optimization, while object size follows a uniform random distribution within a specified interval, and the object position is intra-image optimal. We show that our method improves the segmentation quality when used for training state-of-the-art neural networks on several famous datasets of the SOD field. Combining our method with others surpasses traditional techniques such as horizontal-flip in 0.52% for F-measure and 1.19% for Precision. We also provide an evaluation in 7 different SOD datasets, with 9 distinct evaluation metrics and an average ranking of the evaluated methods.
Eye movements are intricate and dynamic events that contain a wealth of information about the subject and the stimuli. We propose an abstract representation of eye movements that preserve the important nuances in gaze behavior while being stimuli-agnostic. We consider eye movements as raw position and velocity signals and train separate deep temporal convolutional autoencoders. The autoencoders learn micro-scale and macro-scale representations that correspond to the fast and slow features of eye movements. We evaluate the joint representations with a linear classifier fitted on various classification tasks. Our work accurately discriminates between gender and age groups, and outperforms previous works on biometrics and stimuli clasification. Further experiments highlight the validity and generalizability of this method, bringing eye tracking research closer to real-world applications.
This work demonstrates that using the objective with independence assumption for modelling the span probability $P(a_s,a_e) = P(a_s)P(a_e)$ of span starting at position $a_s$ and ending at position $a_e$ has adverse effects. Therefore we propose multiple approaches to modelling joint probability $P(a_s,a_e)$ directly. Among those, we propose a compound objective, composed from the joint probability while still keeping the objective with independence assumption as an auxiliary objective. We find that the compound objective is consistently superior or equal to other assumptions in exact match. Additionally, we identified common errors caused by the assumption of independence and manually checked the counterpart predictions, demonstrating the impact of the compound objective on the real examples. Our findings are supported via experiments with three extractive QA models (BIDAF, BERT, ALBERT) over six datasets and our code, individual results and manual analysis are available online.
In this paper, we focus on the semantic image synthesis task that aims at transferring semantic label maps to photo-realistic images. Existing methods lack effective semantic constraints to preserve the semantic information and ignore the structural correlations in both spatial and channel dimensions, leading to unsatisfactory blurry and artifact-prone results. To address these limitations, we propose a novel Dual Attention GAN (DAGAN) to synthesize photo-realistic and semantically-consistent images with fine details from the input layouts without imposing extra training overhead or modifying the network architectures of existing methods. We also propose two novel modules, i.e., position-wise Spatial Attention Module (SAM) and scale-wise Channel Attention Module (CAM), to capture semantic structure attention in spatial and channel dimensions, respectively. Specifically, SAM selectively correlates the pixels at each position by a spatial attention map, leading to pixels with the same semantic label being related to each other regardless of their spatial distances. Meanwhile, CAM selectively emphasizes the scale-wise features at each channel by a channel attention map, which integrates associated features among all channel maps regardless of their scales. We finally sum the outputs of SAM and CAM to further improve feature representation. Extensive experiments on four challenging datasets show that DAGAN achieves remarkably better results than state-of-the-art methods, while using fewer model parameters. The source code and trained models are available at https://github.com/Ha0Tang/DAGAN.
Majority of the existing graph neural networks (GNN) learn node embeddings that encode their local neighborhoods but not their positions. Consequently, two nodes that are vastly distant but located in similar local neighborhoods map to similar embeddings in those networks. This limitation prevents accurate performance in predictive tasks that rely on position information. In this paper, we develop GraphReach, a position-aware inductive GNN that captures the global positions of nodes through reachability estimations with respect to a set of anchor nodes. The anchors are strategically selected so that reachability estimations across all the nodes are maximized. We show that this combinatorial anchor selection problem is NP-hard and, consequently, develop a greedy (1-1/e) approximation heuristic. Empirical evaluation against state-of-the-art GNN architectures reveal that GraphReach provides up to 40% relative improvement in accuracy. In addition, it is more robust to adversarial attacks.
Besides position bias, which has been well-studied, trust bias is another type of bias prevalent in user interactions with rankings: users are more likely to click incorrectly w.r.t. their preferences on highly ranked items because they trust the ranking system. While previous work has observed this behavior in users, we prove that existing Counterfactual Learning to Rank (CLTR) methods do not remove this bias, including methods specifically designed to mitigate this type of bias. Moreover, we prove that Inverse Propensity Scoring (IPS) is principally unable to correct for trust bias under non-trivial circumstances. Our main contribution is a new estimator based on affine corrections: it both reweights clicks and penalizes items displayed on ranks with high trust bias. Our estimator is the first estimator that is proven to remove the effect of both trust bias and position bias. Furthermore, we show that our estimator is a generalization of the existing CLTR framework: if no trust bias is present, it reduces to the original IPS estimator. Our semi-synthetic experiments indicate that by removing the effect of trust bias in addition to position bias, CLTR can approximate the optimal ranking system even closer than previously possible.
Keyword extraction is an important document process that aims at finding a small set of terms that concisely describe a document's topics. The most popular state-of-the-art unsupervised approaches belong to the family of the graph-based methods that build a graph-of-words and use various centrality measures to score the nodes (candidate keywords). In this work, we follow a different path to detect the keywords from a text document by modeling the main distribution of the document's words using local word vector representations. Then, we rank the candidates based on their position in the text and the distance between the corresponding local vectors and the main distribution's center. We confirm the high performance of our approach compared to strong baselines and state-of-the-art unsupervised keyword extraction methods, through an extended experimental study, investigating the properties of the local representations.
We present CROP (Central Roundish Object Painter), which identifies and paints the object at the center of an RGB image. Primarily CROP works for roundish fruits in various illumination conditions, but surprisingly, it could also deal with images of other organic or inorganic materials, or ones by optical and electron microscopes, although CROP was trained solely by 172 images of fruits. The method involves image segmentation by deep learning, and the architecture of the neural network is a deeper version of the original U-Net. This technique could provide us with a means of automatically collecting statistical data of fruit growth in farms. As an example, we describe our experiment of processing 510 time series photos automatically to collect the data on the size and the position of the target fruit. Our trained neural network CROP and the above automatic programs are available on GitHub with user-friendly interface programs.
In dynamic environments, performance of visual SLAM techniques can be impaired by visual features taken from moving objects. One solution is to identify those objects so that their visual features can be removed for localization and mapping. This paper presents a simple and fast pipeline that uses deep neural networks, extended Kalman filters and visual SLAM to improve both localization and mapping in dynamic environments (around 14 fps on a GTX 1080). Results on the dynamic sequences from the TUM dataset using RTAB-Map as visual SLAM suggest that the approach achieves similar localization performance compared to other state-of-the-art methods, while also providing the position of the tracked dynamic objects, a 3D map free of those dynamic objects, better loop closure detection with the whole pipeline able to run on a robot moving at moderate speed.
This paper presents a method that improve state-of-the-art of the concave point detection methods as a first step to segment overlapping objects on images. It is based on the analysis of the curvature of the objects contour. The method has three main steps. First, we pre-process the original image to obtain the value of the curvature on each contour point. Second, we select regions with higher curvature and we apply a recursive algorithm to refine the previous selected regions. Finally, we obtain a concave point from each region based on the analysis of the relative position of their neighbourhood We experimentally demonstrated that a better concave points detection implies a better cluster division. In order to evaluate the quality of the concave point detection algorithm, we constructed a synthetic dataset to simulate overlapping objects, providing the position of the concave points as a ground truth. As a case study, the performance of a well-known application is evaluated, such as the splitting of overlapped cells in images of peripheral blood smears samples of patients with sickle cell anaemia. We used the proposed method to detect the concave points in clusters of cells and then we separate this clusters by ellipse fitting.
Emotion Cause Extraction (ECE) aims to identify emotion causes from a document after annotating the emotion keywords. Some baselines have been proposed to address this problem, such as rule-based, commonsense based and machine learning methods. We show, however, that a simple random selection approach toward ECE that does not require observing the text achieves similar performance compared to the baselines. We utilized only position information relative to the emotion cause to accomplish this goal. Since position information alone without observing the text resulted in higher F-measure, we therefore uncovered a bias in the ECE single genre Sina-news benchmark. Further analysis showed that an imbalance of emotional cause location exists in the benchmark, with a majority of cause clauses immediately preceding the central emotion clause. We examine the bias from a linguistic perspective, and show that high accuracy rate of current state-of-art deep learning models that utilize location information is only evident in datasets that contain such position biases. The accuracy drastically reduced when a dataset with balanced location distribution is introduced. We therefore conclude that it is the innate bias in this benchmark that caused high accuracy rate of these deep learning models in ECE. We hope that the case study in this paper presents both a cautionary lesson, as well as a template for further studies, in interpreting the superior fit of deep learning models without checking for bias.
Counterfactual evaluation can estimate Click-Through-Rate (CTR) differences between ranking systems based on historical interaction data, while mitigating the effect of position bias and item-selection bias. We introduce the novel Logging-Policy Optimization Algorithm (LogOpt), which optimizes the policy for logging data so that the counterfactual estimate has minimal variance. As minimizing variance leads to faster convergence, LogOpt increases the data-efficiency of counterfactual estimation. LogOpt turns the counterfactual approach - which is indifferent to the logging policy - into an online approach, where the algorithm decides what rankings to display. We prove that, as an online evaluation method, LogOpt is unbiased w.r.t. position and item-selection bias, unlike existing interleaving methods. Furthermore, we perform large-scale experiments by simulating comparisons between thousands of rankers. Our results show that while interleaving methods make systematic errors, LogOpt is as efficient as interleaving without being biased.
This paper addresses the task of score following in sheet music given as unprocessed images. While existing work either relies on OMR software to obtain a computer-readable score representation, or crucially relies on prepared sheet image excerpts, we propose the first system that directly performs score following in full-page, completely unprocessed sheet images. Based on incoming audio and a given image of the score, our system directly predicts the most likely position within the page that matches the audio, outperforming current state-of-the-art image-based score followers in terms of alignment precision. We also compare our method to an OMR-based approach and empirically show that it can be a viable alternative to such a system.
Learning-to-Rank (LTR) models trained from implicit feedback (e.g. clicks) suffer from inherent biases. A well-known one is the position bias -- documents in top positions are more likely to receive clicks due in part to their position advantages. To unbiasedly learn to rank, existing counterfactual frameworks first estimate the propensity (probability) of missing clicks with intervention data from a small portion of search traffic, and then use inverse propensity score (IPS) to debias LTR algorithms on the whole data set. These approaches often assume the propensity only depends on the position of the document, which may cause high estimation variance in applications where the search context (e.g. query, user) varies frequently. While context-dependent propensity models reduce variance, accurate estimations may require randomization or intervention on a large amount of traffic, which may not be realistic in real-world systems, especially for long tail queries. In this work, we employ heterogeneous treatment effect estimation techniques to estimate position bias when intervention click data is limited. We then use such estimations to debias the observed click distribution and re-draw a new de-biased data set, which can be used for any LTR algorithms. We conduct simulations with varying experiment conditions and show the effectiveness of the proposed method in regimes with long tail queries and sparse clicks.
Ellipse fitting, an essential component in pupil or iris tracking based video oculography, is performed on previously segmented eye parts generated using various computer vision techniques. Several factors, such as occlusions due to eyelid shape, camera position or eyelashes, frequently break ellipse fitting algorithms that rely on well-defined pupil or iris edge segments. In this work, we propose training a convolutional neural network to directly segment entire elliptical structures and demonstrate that such a framework is robust to occlusions and offers superior pupil and iris tracking performance (at least 10$\%$ and 24$\%$ increase in pupil and iris center detection rate respectively within a two-pixel error margin) compared to using standard eye parts segmentation for multiple publicly available synthetic segmentation datasets.
Convolutional Neural Networks (CNNs) model long-range dependencies by deeply stacking convolution operations with small window sizes, which makes the optimizations difficult. This paper presents region-based non-local (RNL) operations as a family of self-attention mechanisms, which can directly capture long-range dependencies without using a deep stack of local operations. Given an intermediate feature map, our method recalibrates the feature at a position by aggregating the information from the neighboring regions of all positions. By combining a channel attention module with the proposed RNL, we design an attention chain, which can be integrated into the off-the-shelf CNNs for end-to-end training. We evaluate our method on two video classification benchmarks. The experimental results of our method outperform other attention mechanisms, and we achieve state-of-the-art performance on the Something-Something V1 dataset.
The attention-based encoder-decoder framework has recently achieved impressive results for scene text recognition, and many variants have emerged with improvements in recognition quality. However, it performs poorly on contextless texts (e.g., random character sequences) which is unacceptable in most of real application scenarios. In this paper, we first deeply investigate the decoding process of the decoder. We empirically find that a representative character-level sequence decoder utilizes not only context information but also positional information. Contextual information, which the existing approaches heavily rely on, causes the problem of attention drift. To suppress such side-effect, we propose a novel position enhancement branch, and dynamically fuse its outputs with those of the decoder attention module for scene text recognition. Specifically, it contains a position aware module to enable the encoder to output feature vectors encoding their own spatial positions, and an attention module to estimate glimpses using the positional clue (i.e., the current decoding time step) only. The dynamic fusion is conducted for more robust feature via an element-wise gate mechanism. Theoretically, our proposed method, dubbed \emph{RobustScanner}, decodes individual characters with dynamic ratio between context and positional clues, and utilizes more positional ones when the decoding sequences with scarce context, and thus is robust and practical. Empirically, it has achieved new state-of-the-art results on popular regular and irregular text recognition benchmarks while without much performance drop on contextless benchmarks, validating its robustness in both contextual and contextless application scenarios.
Much of the recently available research and challenge data lack the meta-data containing any information about the patient position. This paper presents a tool for automatic rotation of CT data into a standardized (HFS) patient position. The proposed method is based on the prediction of rotation angle with CNN, and it achieved nearly perfect results with an accuracy of 99.55 %. We provide implementations with easy to use an example for both Matlab and Python (PyTorch), which can be used, for example, for automatic rotation correction of VerSe2020 challenge data.
Monocular multi-object detection and localization in 3D space has been proven to be a challenging task. The MoNet3D algorithm is a novel and effective framework that can predict the 3D position of each object in a monocular image and draw a 3D bounding box for each object. The MoNet3D method incorporates prior knowledge of the spatial geometric correlation of neighbouring objects into the deep neural network training process to improve the accuracy of 3D object localization. Experiments on the KITTI dataset show that the accuracy for predicting the depth and horizontal coordinates of objects in 3D space can reach 96.25\% and 94.74\%, respectively. Moreover, the method can realize the real-time image processing at 27.85 FPS, showing promising potential for embedded advanced driving-assistance system applications. Our code is publicly available at https://github.com/CQUlearningsystemgroup/YicongPeng.