Abstract: Image values are the result of a combination of visual information coming from multiple sources. Recovering information from the multiple factors thatproduced an image seems a hard and ill-posed problem. However, it is important to observe that humans develop the ability to interpret images and recognize and isolate specific physical properties of the scene.
Images describing a single physical characteristic of an scene are called intrinsic images. These images would benefit most computer vision tasks which are often affected by the multiple complex effects that are usually found in natural images (e.g. cast shadows, specularities, interreflections...).
In this thesis we analyze the problem of intrinsic image estimation from different perspectives, including the theoretical formulation of the problem, the visual cues that can be used to estimate the intrinsic components and the evaluation mechanisms of the problem.

Abstract: Handwritten text recognition in low resource scenarios, such as manuscripts with rare alphabets, is a challenging problem. In this paper, we propose a few-shot learning-based handwriting recognition approach that significantly reduces the human annotation process, by requiring only a few images of each alphabet symbols. The method consists of detecting all the symbols of a given alphabet in a textline image and decoding the obtained similarity scores to the final sequence of transcribed symbols. Our model is first pretrained on synthetic line images generated from an alphabet, which could differ from the alphabet of the target domain. A second training step is then applied to reduce the gap between the source and the target data. Since this retraining would require annotation of thousands of handwritten symbols together with their bounding boxes, we propose to avoid such human effort through an unsupervised progressive learning approach that automatically assigns pseudo-labels to the unlabeled data. The evaluation on different datasets shows that our model can lead to competitive results with a significant reduction in human effort. The code will be publicly available in the following repository: https://github.com/dali92002/HTRbyMatching

Abstract: Bronchoscopy biopsy can be used to diagnose lung cancer without risking complications of other interventions like transthoracic needle aspiration. During bronchoscopy, the clinician has to navigate through the bronchial tree to the target lesion. A main drawback is the difficulty to check whether the exploration is following the correct path. The usual guidance using fluoroscopy implies repeated radiation of the clinician, while alternative systems (like electromagnetic navigation) require specific equipment that increases intervention costs. We propose to compute the navigated path using anatomical landmarks extracted from the sole analysis of videobronchoscopy images. Such landmarks allow matching the current exploration to the path previously planned on a CT to indicate clinician whether the planning is being correctly followed or not. We present a feasibility study of our landmark based CT-video matching using bronchoscopic videos simulated on a virtual bronchoscopy interactive interface.

Abstract: Deep video action recognition models have been highly successful in recent years but require large quantities of manually-annotated data, which are expensive and laborious to obtain. In this work, we investigate the generation of synthetic training data for video action recognition, as synthetic data have been successfully used to supervise models for a variety of other computer vision tasks. We propose an interpretable parametric generative model of human action videos that relies on procedural generation, physics models and other components of modern game engines. With this model we generate a diverse, realistic, and physically plausible dataset of human action videos, called PHAV for “Procedural Human Action Videos”. PHAV contains a total of 39,982 videos, with more than 1000 examples for each of 35 action categories. Our video generation approach is not limited to existing motion capture sequences: 14 of these 35 categories are procedurally-defined synthetic actions. In addition, each video is represented with 6 different data modalities, including RGB, optical flow and pixel-level semantic labels. These modalities are generated almost simultaneously using the Multiple Render Targets feature of modern GPUs. In order to leverage PHAV, we introduce a deep multi-task (i.e. that considers action classes from multiple datasets) representation learning architecture that is able to simultaneously learn from synthetic and real video datasets, even when their action categories differ. Our experiments on the UCF-101 and HMDB-51 benchmarks suggest that combining our large set of synthetic videos with small real-world datasets can boost recognition performance. Our approach also significantly outperforms video representations produced by fine-tuning state-of-the-art unsupervised generative models of videos.

Abstract: Documenting the severity of a static or dynamic Central Airway Obstruction (CAO) is crucial to establish proper diagnosis and treatment, predict possible treatment effects and better follow-up the patients. The subjective visual evaluation of a stenosis during video-bronchoscopy still remains the most common way to assess a CAO in spite of a consensus among experts for a need to standardize all calculations [1].
The Computer Vision Center in cooperation with the «Hospital de Bellvitge», has developed a System for Endoscopic Stenosis Assessment (SENSA), which computes CAO directly by analyzing standard bronchoscopic data without the need of using other imaging tecnologies.