This paper tackles the problem of novel category discovery (NCD), which aims to discriminate unknown categories in large-scale image collections. The NCD task is challenging due to the closeness to the real-world scenarios, where we have only encountered some partial classes and images. Unlike other works on the NCD, we leverage the prototypes to emphasize the importance of category discrimination and alleviate the issue of missing annotations of novel classes. Concretely, we propose a novel adaptive prototype learning method consisting of two main stages: prototypical representation learning and prototypical self-training. In the first stage, we obtain a robust feature extractor, which could serve for all images with base and novel categories. This ability of instance and category discrimination of the feature extractor is boosted by self-supervised learning and adaptive prototypes. In the second stage, we utilize the prototypes again to rectify offline pseudo labels and train a final parametric classifier for category clustering. We conduct extensive experiments on four benchmark datasets and demonstrate the effectiveness and robustness of the proposed method with state-of-the-art performance.
Continuous monitoring of foot ulcer healing is needed to ensure the efficacy of a given treatment and to avoid any possibility of deterioration. Foot ulcer segmentation is an essential step in wound diagnosis. We developed a model that is similar in spirit to the well-established encoder-decoder and residual convolution neural networks. Our model includes a residual connection along with a channel and spatial attention integrated within each convolution block. A simple patch-based approach for model training, test time augmentations, and majority voting on the obtained predictions resulted in superior performance. Our model did not leverage any readily available backbone architecture, pre-training on a similar external dataset, or any of the transfer learning techniques. The total number of network parameters being around 5 million made it a significantly lightweight model as compared with the available state-of-the-art models used for the foot ulcer segmentation task. Our experiments presented results at the patch-level and image-level. Applied on publicly available Foot Ulcer Segmentation (FUSeg) Challenge dataset from MICCAI 2021, our model achieved state-of-the-art image-level performance of 88.22% in terms of Dice similarity score and ranked second in the official challenge leaderboard. We also showed an extremely simple solution that could be compared against the more advanced architectures.
The optics of any camera degrades the sharpness of photographs, which is a key visual quality criterion. This degradation is characterized by the point-spread function (PSF), which depends on the wavelengths of light and is variable across the imaging field. In this paper, we propose a two-step scheme to correct optical aberrations in a single raw or JPEG image, i.e., without any prior information on the camera or lens. First, we estimate local Gaussian blur kernels for overlapping patches and sharpen them with a non-blind deblurring technique. Based on the measurements of the PSFs of dozens of lenses, these blur kernels are modeled as RGB Gaussians defined by seven parameters. Second, we remove the remaining lateral chromatic aberrations (not contemplated in the first step) with a convolutional neural network, trained to minimize the red/green and blue/green residual images. Experiments on both synthetic and real images show that the combination of these two stages yields a fast state-of-the-art blind optical aberration compensation technique that competes with commercial non-blind algorithms.
Pixel-level 2D object semantic understanding is an important topic in computer vision and could help machine deeply understand objects (e.g. functionality and affordance) in our daily life. However, most previous methods directly train on correspondences in 2D images, which is end-to-end but loses plenty of information in 3D spaces. In this paper, we propose a new method on predicting image corresponding semantics in 3D domain and then projecting them back onto 2D images to achieve pixel-level understanding. In order to obtain reliable 3D semantic labels that are absent in current image datasets, we build a large scale keypoint knowledge engine called KeypointNet, which contains 103,450 keypoints and 8,234 3D models from 16 object categories. Our method leverages the advantages in 3D vision and can explicitly reason about objects self-occlusion and visibility. We show that our method gives comparative and even superior results on standard semantic benchmarks.
Eye-in-hand camera calibration is a fundamental and long-studied problem in robotics. We present a study on using learning-based methods for solving this problem online from a single RGB image, whilst training our models with entirely synthetic data. We study three main approaches: one direct regression model that directly predicts the extrinsic matrix from an image, one sparse correspondence model that regresses 2D keypoints and then uses PnP, and one dense correspondence model that uses regressed depth and segmentation maps to enable ICP pose estimation. In our experiments, we benchmark these methods against each other and against well-established classical methods, to find the surprising result that direct regression outperforms other approaches, and we perform noise-sensitivity analysis to gain further insights into these results.
Supervised learning is well-known to fail at generalization under distribution shifts. In typical clinical settings, the source data is inaccessible and the target distribution is represented with a handful of samples: adaptation can only happen at test time on a few or even a single subject(s). We investigate test-time single-subject adaptation for segmentation, and propose a Shape-guided Entropy Minimization objective for tackling this task. During inference for a single testing subject, our loss is minimized with respect to the batch normalization's scale and bias parameters. We show the potential of integrating various shape priors to guide adaptation to plausible solutions, and validate our method in two challenging scenarios: MRI-to-CT adaptation of cardiac segmentation and cross-site adaptation of prostate segmentation. Our approach exhibits substantially better performances than the existing test-time adaptation methods. Even more surprisingly, it fares better than state-of-the-art domain adaptation methods, although it forgoes training on additional target data during adaptation. Our results question the usefulness of training on target data in segmentation adaptation, and points to the substantial effect of shape priors on test-time inference. Our framework can be readily used for integrating various priors and for adapting any segmentation network, and our code is available.
Imbalanced data pose challenges for deep learning based classification models. One of the most widely-used approaches for tackling imbalanced data is re-weighting, where training samples are associated with different weights in the loss function. Most of existing re-weighting approaches treat the example weights as the learnable parameter and optimize the weights on the meta set, entailing expensive bilevel optimization. In this paper, we propose a novel re-weighting method based on optimal transport (OT) from a distributional point of view. Specifically, we view the training set as an imbalanced distribution over its samples, which is transported by OT to a balanced distribution obtained from the meta set. The weights of the training samples are the probability mass of the imbalanced distribution and learned by minimizing the OT distance between the two distributions. Compared with existing methods, our proposed one disengages the dependence of the weight learning on the concerned classifier at each iteration. Experiments on image, text and point cloud datasets demonstrate that our proposed re-weighting method has excellent performance, achieving state-of-the-art results in many cases and providing a promising tool for addressing the imbalanced classification issue.
Analysing 88 sources published from 2011 to 2021, this paper presents a first systematic review of the computer vision-based analysis of buildings and the built environments to assess its value to architectural and urban design studies. Following a multi-stage selection process, the types of algorithms and data sources used are discussed in respect to architectural applications such as a building classification, detail classification, qualitative environmental analysis, building condition survey, and building value estimation. This reveals current research gaps and trends, and highlights two main categories of research aims. First, to use or optimise computer vision methods for architectural image data, which can then help automate time-consuming, labour-intensive, or complex tasks of visual analysis. Second, to explore the methodological benefits of machine learning approaches to investigate new questions about the built environment by finding patterns and relationships between visual, statistical, and qualitative data, which can overcome limitations of conventional manual analysis. The growing body of research offers new methods to architectural and design studies, with the paper identifying future challenges and directions of research.
This work presents an unsupervised deep discriminant analysis for clustering. The method is based on deep neural networks and aims to minimize the intra-cluster discrepancy and maximize the inter-cluster discrepancy in an unsupervised manner. The method is able to project the data into a nonlinear low-dimensional latent space with compact and distinct distribution patterns such that the data clusters can be effectively identified. We further provide an extension of the method such that available graph information can be effectively exploited to improve the clustering performance. Extensive numerical results on image and non-image data with or without graph information demonstrate the effectiveness of the proposed methods.
Extraterrestrial rovers with a general-purpose robotic arm have many potential applications in lunar and planetary exploration. Introducing autonomy into such systems is desirable for increasing the time that rovers can spend gathering scientific data and collecting samples. This work investigates the applicability of deep reinforcement learning for vision-based robotic grasping of objects on the Moon. A novel simulation environment with procedurally-generated datasets is created to train agents under challenging conditions in unstructured scenes with uneven terrain and harsh illumination. A model-free off-policy actor-critic algorithm is then employed for end-to-end learning of a policy that directly maps compact octree observations to continuous actions in Cartesian space. Experimental evaluation indicates that 3D data representations enable more effective learning of manipulation skills when compared to traditionally used image-based observations. Domain randomization improves the generalization of learned policies to novel scenes with previously unseen objects and different illumination conditions. To this end, we demonstrate zero-shot sim-to-real transfer by evaluating trained agents on a real robot in a Moon-analogue facility.