Abstract

It is critical for aerial robots flying in city scale urban environments to make very quick estimates of a building depth with respect to itself. It should be done in a matter of few views to navigate itself, avoiding collisions with such a towering structure. As such, no one has attacked this problem. We bring together several modules combining deep learning and 3D vision to showcase a quick reconstruction in a few views. We exploit the inherent planar structure in the buildings (facades, windows) for this purpose. We evaluate the efficacy of our pipeline with various constraints and errors from multi-view geometry using ablation studies. We then retrieve the skyline of the buildings in synthetic as well as real-world scenes.

Abstract

Surface normal estimation is an essential component of several computer and robot vision pipelines. While this problem has been extensively studied, most approaches are geared towards indoor scenes and often rely on multiple modalities (depth, multiple views) for accurate estimation of normal maps. Outdoor scenes pose a greater challenge as they exhibit significant lighting variation, often contain occluders, and structures like building facades are often ridden with numerous windows and protrusions. Conventional supervised learning schemes excel in indoor scenes, but do not exhibit competitive performance when trained and deployed in outdoor environments. Furthermore, they involve complex network architectures and require many more trainable parameters. To tackle these challenges, we present an adversarial learning scheme that regularizes the output normal maps from a neural network to appear more realistic, by using a small number of precisely annotated examples. Our method presents a lightweight and simpler architecture, while improving performance by at least 1.5x across most metrics. We evaluate our approaches against the state-of-the-art on normal map estimation, on a synthetic and a real outdoor dataset, and observe significant performance enhancements.

Abstract

Histopathology image analysis is widely accepted as a gold standard for cancer diagnosis. The Cancer Genome Atlas (TCGA) contains large repositories of histopathology whole slide images spanning across several organs and subtypes. However, not much work has gone into analysis of all the organs and subtypes and their similarities. Our work attempts to bridge this gap by training deep learning models to classify cancer vs normal patches for 11 subtypes spanning 7 organs (9792 tissue slides) to achieve near-perfect classification performance. We used these models to investigate their performances in the test set of other organs (cross organ inference). We found that every model had a good cross organ inference accuracy when tested on breast, colorectal and liver cancers. Further, a high accuracy is observed between models trained on the cancer subtypes originating from same organ (kidney and lung). We also validated these performances by showing the separability of cancer and normal samples in a high dimensional feature space. We further hypothesized that the high cross organ inferences are due to shared tumor morphologies among organs. We validated the hypothesis by showing the overlap in the Gradient-weighted Class Activation Mapping (GradCAM) visualizations and similarities in the distributions of nuclei geometrical features present within the high attention regions. Keywords: TCGA, histopathology, cancer classification, cross-organ inference, deep learning, tissue morphology, class activation map.

Abstract

Owing to the large dimensions of the histopathology whole slide images (WSI), visually searching for clinically significant regions (patches) is a tedious task for a medical expert. Sequential analysis of several such images further increases the workload resulting in poor diagnosis. A major impediment towards automating this task using deep learning models is that it requires a huge chunk of laboriously annotated data in the form of WSI patches. Our work suggests a novel CNN-based, expert feedback-driven interactive learning technique to mitigate this issue. The proposed method seeks to acquire labels of the most informative patches in small increments with multiple feedback rounds to maximize the throughput. It requires the expert to query a patch of interest from one slide and provide feedback to a set of unlabelled patches chosen using the proposed sampling strategy from a ranked list. The experiments on a large patient cohort of colorectal cancer histological patches (100K images with nine classes of tissues) show a significant reduction (≈ 95%) in the amount of labelled data required to achieve state-of theart results when compared to other existing interactive learning methods (35% − 50%). We also demonstrate the utility of the proposed technique to assist a WSI tumor segmentation annotation task using the ICIAR breast cancer challenge dataset (≈ 12.5K patches per slide). The proposed technique reduces the scanning and searching area to about 2% of the total area of WSI (by seeking labels of ≈ 250 informative patches only) and achieves segmentation outputs with 85% IOU. Thus our work helps avoid the routine procedure of exhaustive scanning and searching during annotation and diagnosis in general.

Abstract

Digitizing via scanning the physical artifact often forms the first primary step in preserving historical handwritten manuscripts. To maximally utilize scanner surface area and minimize manual labor, multiple manuscripts are usually scanned together into a scanned image. Therefore, the first crucial task in manuscript content understanding is to ensure that each of the individual manuscripts within a scanned image can be isolated (segmented) on a per-instance basis. Existing deep network based approaches for manuscript layout understanding implicitly assume a single or two manuscripts per image. Since this assumption may be routinely violated, there is a need for a precursor system which extracts individual manuscripts before downstream processing. Another challenge is the highly curved and deformed boundaries of manuscripts, causing them to often overlap with each other. To tackle such challenges, we introduce a new document image dataset called IMMI (Indic Multi Manuscript Images). To improve the efficiency of dataset and aid deep network training, we also propose an approach which generates synthetic images to augment sourced non-synthetic images. We conduct experiments using modified versions of existing document instance segmentation frameworks. The results demonstrate the efficacy of the new frameworks for the task. Overall, our contributions enable robust extraction of individual historical manuscript pages. This in turn, could potentially enable better performance on downstream tasks such as region-level instance segmentation within handwritten manuscripts and optical character recognition

Abstract

Given a monocular color image of a warehouse rack, we aim to predict the bird's-eye view layout for each shelf in the rack, which we term as' multi-layer'layout prediction. To this end, we present RackLay, a deep neural network for real-time shelf layout estimation from a single image. Unlike previous layout estimation methods which provide a single layout for the dominant ground plane alone, RackLay estimates the top-viewand front-view layout for each shelf in the considered rack populated with objects. RackLay's architecture and its variants are versatile and estimate accurate layouts for diverse scenes characterized by varying number of visible shelves in an image, large range in shelf occupancy factor and varied background clutter. Given the extreme paucity of datasets in this space and the difficulty involved in acquiring real data from warehouses, we additionally release a flexible synthetic dataset generation …

Abstract

Assessing the number of street trees is essential for evaluating urban greenery and can help municipalities employ solutions to identify tree-starved streets. It can also help identify roads with different levels of deforestation and afforestation over time. Yet, there has been little work in the area of street trees quantification. This work first explains a data collection setup carefully designed for counting roadside trees. We then describe a unique annotation procedure aimed at robustly detecting and quantifying trees. We work on a dataset of around 1300 Indian road scenes annotated with over 2500 street trees. We additionally use the five held-out videos covering 25 km of roads for counting trees. We finally propose a street tree detection, counting, and visualization framework using current object detectors and a novel yet simple counting algorithm owing to the thoughtful collection setup. We find that the high-level …

Abstract

Sentiment analysis aims to detect the overall sentiment, i.e., the polarity of a sentence, paragraph, or text span, without considering the entities mentioned and their aspects. Aspectbased sentiment analysis aims to extract the aspects of the given target entities and their respective sentiments. Prior works formulate this as a sequence tagging problem or solve this task using a span-based extract-thenclassify framework where first all the opinion targets are extracted from the sentence, and then with the help of span representations, the targets are classified as positive, negative, or neutral. The sequence tagging problem suffers from issues like sentiment inconsistency and colossal search space. Whereas, Span-based extract-then-classify framework suffers from issues such as half-word coverage and overlapping spans. To overcome this, we propose a similar span-based extract-then-classify framework with a novel and improved heuristic. Experiments on the three benchmark datasets (Restaurant14, Laptop14, Restaurant15) show our model consistently outperforms the current state-of-the-art. Moreover, we also present a novel supervised movie reviews dataset (Movie20) and a pseudo-labeled movie reviews dataset (moviesLarge) made explicitly for this task in English language and report the results on the novel Movie20 dataset as well.

Abstract

Human-robot interaction, particularly in wheeled mobile robots that can autonomously assist humans to traverse dynamically changing environments is a field of active research. Integrated motion planning and obstacle-avoidance pose a considerable challenge for an autonomous person-following robot (PFR). And, scenarios with intersections and occlusions along the path only increase the complexity in sustained tracking. In this paper, we use model predictive control (MPC) with earlyrelocation (ER) strategy to formulate a prioritized tracking scheme and implement it for a differential-drive system. Our approach ensures that the target person stays within the field of view (FOV) of the PFR consistently, even while it maneuvers intersections or crowded spots, by adding new locations to its updated path. As trajectory generation in such cases must be incremental to accommodate new information, the use of efficient representations is key. To that end, we build this social representation of following a person directly into the controller itself. MPC can naturally handle such state and input limitations as constraints to solve an on-line optimization at each time step. A non-linear MPC with ER is thus devised and tested with increasing levels of complexity arising from occlusions due to the map and its dynamic actors. By using 2D simulations, we show that for slow and medium walking speeds of the target person, the controller can plan maneuvers with an adequate margin of over 20 Hz apt for achieving a near real-time personfollowing behaviour.

Abstract

This paper presents the design of an in-pipe climbing robot that operates using a novel ‘Three-output open differential’(3-OOD) mechanism to traverse complex networks of pipes. Conventional wheeled/tracked in-pipe climbing robots are prone to slip and drag while traversing in pipe bends. The 3-OOD mechanism helps in achieving the novel result of eliminating slip and drag in the robot tracks during motion. The proposed differential realizes the functional abilities of the traditional two-output differential, which is achieved the first time for a differential with three outputs. The 3-OOD mechanism mechanically modulates the track speeds of the robot based on the forces exerted on each track inside the pipe network, by eliminating the need for any active control. The simulation of the robot traversing in the pipe network in different orientations and in pipe-bends without slip shows the proposed design’s effectiveness.