Loong-Fah Cheong

Aashish Sharma, Robby T. Tan, Loong-Fah Cheong

Most existing methods for CRF estimation from a single image fail to handle general real images. For instance, EdgeCRF based on colour patches extracted from edges works effectively only when the presence of noise is insignificant, which is not the case for many real images; and, CRFNet, a recent method based on fully supervised deep learning works only for the CRFs that are in the training data, and hence fail to deal with other possible CRFs beyond the training data. To address these problems, we introduce a non-deep-learning method using prediction consistency and gradual refinement. First, we rely more on the patches of the input image that provide more consistent predictions. If the predictions from a patch are more consistent, it means that the patch is likely to be less affected by noise or any inferior colour combinations, and hence, it can be more reliable for CRF estimation. Second, we employ a gradual refinement scheme in which we start from a simple CRF model to generate a result which is more robust to noise but less accurate, and then we gradually increase the model's complexity to improve the result. This is because a simple model, while being less accurate, overfits less to noise than a complex model does. Our experiments show that our method outperforms the existing single-image methods for daytime and nighttime real images. We further propose a more efficient deep learning extension that performs test-time training (based on unsupervised losses) on the test input image. This provides our method better generalization performance than CRFNet making it more practically applicable for CRF estimation for general real images.

Aashish Sharma, Lionel Heng, Loong-Fah Cheong et al.

Nighttime stereo depth estimation is still challenging, as assumptions associated with daytime lighting conditions do not hold any longer. Nighttime is not only about low-light and dense noise, but also about glow/glare, flares, non-uniform distribution of light, etc. One of the possible solutions is to train a network on night stereo images in a fully supervised manner. However, to obtain proper disparity ground-truths that are dense, independent from glare/glow, and have sufficiently far depth ranges is extremely intractable. To address the problem, we introduce a network joining day/night translation and stereo. In training the network, our method does not require ground-truth disparities of the night images, or paired day/night images. We utilize a translation network that can render realistic night stereo images from day stereo images. We then train a stereo network on the rendered night stereo images using the available disparity supervision from the corresponding day stereo images, and simultaneously also train the day/night translation network. We handle the fake depth problem, which occurs due to the unsupervised/unpaired translation, for light effects (e.g., glow/glare) and uninformative regions (e.g., low-light and saturated regions), by adding structure-preservation and weighted-smoothness constraints. Our experiments show that our method outperforms the baseline methods on night images.

Xun Xu, Loong-Fah Cheong, Zhuwen Li

Many real-world video sequences cannot be conveniently categorized as general or degenerate; in such cases, imposing a false dichotomy in using the fundamental matrix or homography model for motion segmentation on video sequences would lead to difficulty. Even when we are confronted with a general scene-motion, the fundamental matrix approach as a model for motion segmentation still suffers from several defects, which we discuss in this paper. The full potential of the fundamental matrix approach could only be realized if we judiciously harness information from the simpler homography model. From these considerations, we propose a multi-model spectral clustering framework that synergistically combines multiple models (homography and fundamental matrix) together. We show that the performance can be substantially improved in this way. For general motion segmentation tasks, the number of independently moving objects is often unknown a priori and needs to be estimated from the observations. This is referred to as model selection and it is essentially still an open research problem. In this work, we propose a set of model selection criteria balancing data fidelity and model complexity. We perform extensive testing on existing motion segmentation datasets with both segmentation and model selection tasks, achieving state-of-the-art performance on all of them; we also put forth a more realistic and challenging dataset adapted from the KITTI benchmark, containing real-world effects such as strong perspectives and strong forward translations not seen in the traditional datasets.

Ahmad W. Bitar, Jean-Philippe Ovarlez, Loong-Fah Cheong et al.

In this work, a novel target detector for hyperspectral imagery is developed. The detector is independent on the unknown covariance matrix, behaves well in large dimensions, distributional free, invariant to atmospheric effects, and does not require a background dictionary to be constructed. Based on a modification of the robust principal component analysis (RPCA), a given hyperspectral image (HSI) is regarded as being made up of the sum of a low-rank background HSI and a sparse target HSI that contains the targets based on a pre-learned target dictionary specified by the user. The sparse component is directly used for the detection, that is, the targets are simply detected at the non-zero entries of the sparse target HSI. Hence, a novel target detector is developed, which is simply a sparse HSI generated automatically from the original HSI, but containing only the targets with the background is suppressed. The detector is evaluated on real experiments, and the results of which demonstrate its effectiveness for hyperspectral target detection especially when the targets are well matched to the surroundings.

Xun Xu, Loong-Fah Cheong, Zhuwen Li

Subspace clustering has been extensively studied from the hypothesis-and-test, algebraic, and spectral clustering based perspectives. Most assume that only a single type/class of subspace is present. Generalizations to multiple types are non-trivial, plagued by challenges such as choice of types and numbers of models, sampling imbalance and parameter tuning. In this work, we formulate the multi-type subspace clustering problem as one of learning non-linear subspace filters via deep multi-layer perceptrons (mlps). The response to the learnt subspace filters serve as the feature embedding that is clustering-friendly, i.e., points of the same clusters will be embedded closer together through the network. For inference, we apply K-means to the network output to cluster the data. Experiments are carried out on both synthetic and real world multi-type fitting problems, producing state-of-the-art results.

Bingbing Zhuang, Loong-Fah Cheong, Gim Hee Lee

In this paper, we develop a modified differential Structure from Motion (SfM) algorithm that can estimate relative pose from two consecutive frames despite of Rolling Shutter (RS) artifacts. In particular, we show that under constant velocity assumption, the errors induced by the rolling shutter effect can be easily rectified by a linear scaling operation on each optical flow. We further propose a 9-point algorithm to recover the relative pose of a rolling shutter camera that undergoes constant acceleration motion. We demonstrate that the dense depth maps recovered from the relative pose of the RS camera can be used in a RS-aware warping for image rectification to recover high-quality Global Shutter (GS) images. Experiments on both synthetic and real RS images show that our RS-aware differential SfM algorithm produces more accurate results on relative pose estimation and 3D reconstruction from images distorted by RS effect compared to standard SfM algorithms that assume a GS camera model. We also demonstrate that our RS-aware warping for image rectification method outperforms state-of-the-art commercial software products, i.e. Adobe After Effects and Apple Imovie, at removing RS artifacts.

Xun Xu, Loong-Fah Cheong, Zhuwen Li

Multi-model fitting has been extensively studied from the random sampling and clustering perspectives. Most assume that only a single type/class of model is present and their generalizations to fitting multiple types of models/structures simultaneously are non-trivial. The inherent challenges include choice of types and numbers of models, sampling imbalance and parameter tuning, all of which render conventional approaches ineffective. In this work, we formulate the multi-model multi-type fitting problem as one of learning deep feature embedding that is clustering-friendly. In other words, points of the same clusters are embedded closer together through the network. For inference, we apply K-means to cluster the data in the embedded feature space and model selection is enabled by analyzing the K-means residuals. Experiments are carried out on both synthetic and real world multi-type fitting datasets, producing state-of-the-art results. Comparisons are also made on single-type multi-model fitting tasks with promising results as well.

Bingbing Zhuang, Loong-Fah Cheong, Gim Hee Lee

Many existing translation averaging algorithms are either sensitive to disparate camera baselines and have to rely on extensive preprocessing to improve the observed Epipolar Geometry graph, or if they are robust against disparate camera baselines, require complicated optimization to minimize the highly nonlinear angular error objective. In this paper, we carefully design a simple yet effective bilinear objective function, introducing a variable to perform the requisite normalization. The objective function enjoys the baseline-insensitive property of the angular error and yet is amenable to simple and efficient optimization by block coordinate descent, with good empirical performance. A rotation-assisted Iterative Reweighted Least Squares scheme is further put forth to help deal with outliers. We also contribute towards a better understanding of the behavior of two recent convex algorithms, LUD and Shapefit/kick, clarifying the underlying subtle difference that leads to the performance gap. Finally, we demonstrate that our algorithm achieves overall superior accuracies in benchmark dataset compared to state-of-theart methods, and is also several times faster.

Xun Xu, Loong-Fah Cheong, Zhuwen Li

Many real-world sequences cannot be conveniently categorized as general or degenerate; in such cases, imposing a false dichotomy in using the fundamental matrix or homography model for motion segmentation would lead to difficulty. Even when we are confronted with a general scene-motion, the fundamental matrix approach as a model for motion segmentation still suffers from several defects, which we discuss in this paper. The full potential of the fundamental matrix approach could only be realized if we judiciously harness information from the simpler homography model. From these considerations, we propose a multi-view spectral clustering framework that synergistically combines multiple models together. We show that the performance can be substantially improved in this way. We perform extensive testing on existing motion segmentation datasets, achieving state-of-the-art performance on all of them; we also put forth a more realistic and challenging dataset adapted from the KITTI benchmark, containing real-world effects such as strong perspectives and strong forward translations not seen in the traditional datasets.

Ruoteng Li, Loong-Fah Cheong, Robby T. Tan

Given a single input rainy image, our goal is to visually remove rain streaks and the veiling effect caused by scattering and transmission of rain streaks and rain droplets. We are particularly concerned with heavy rain, where rain streaks of various sizes and directions can overlap each other and the veiling effect reduces contrast severely. To achieve our goal, we introduce a scale-aware multi-stage convolutional neural network. Our main idea here is that different sizes of rain-streaks visually degrade the scene in different ways. Large nearby streaks obstruct larger regions and are likely to reflect specular highlights more prominently than smaller distant streaks. These different effects of different streaks have their own characteristics in their image features, and thus need to be treated differently. To realize this, we create parallel sub-networks that are trained and made aware of these different scales of rain streaks. To our knowledge, this idea of parallel sub-networks that treats the same class of objects according to their unique sub-classes is novel, particularly in the context of rain removal. To verify our idea, we conducted experiments on both synthetic and real images, and found that our method is effective and outperforms the state-of-the-art methods.

An Tran, Loong-Fah Cheong

This paper proposes a two-stream flow-guided convolutional attention networks for action recognition in videos. The central idea is that optical flows, when properly compensated for the camera motion, can be used to guide attention to the human foreground. We thus develop cross-link layers from the temporal network (trained on flows) to the spatial network (trained on RGB frames). These cross-link layers guide the spatial-stream to pay more attention to the human foreground areas and be less affected by background clutter. We obtain promising performances with our approach on the UCF101, HMDB51 and Hollywood2 datasets.

Ruoteng Li, Robby T. Tan, Loong-Fah Cheong

Optical flow estimation in the rainy scenes is challenging due to background degradation introduced by rain streaks and rain accumulation effects in the scene. Rain accumulation effect refers to poor visibility of remote objects due to the intense rainfall. Most existing optical flow methods are erroneous when applied to rain sequences because the conventional brightness constancy constraint (BCC) and gradient constancy constraint (GCC) generally break down in this situation. Based on the observation that the RGB color channels receive raindrop radiance equally, we introduce a residue channel as a new data constraint to reduce the effect of rain streaks. To handle rain accumulation, our method decomposes the image into a piecewise-smooth background layer and a high-frequency detail layer. It also enforces the BCC on the background layer only. Results on both synthetic dataset and real images show that our algorithm outperforms existing methods on different types of rain sequences. To our knowledge, this is the first optical flow method specifically dealing with rain.

Yu-Xiang Wang, Choon Meng Lee, Loong-Fah Cheong et al.

Low-rank matrix completion is a problem of immense practical importance. Recent works on the subject often use nuclear norm as a convex surrogate of the rank function. Despite its solid theoretical foundation, the convex version of the problem often fails to work satisfactorily in real-life applications. Real data often suffer from very few observations, with support not meeting the random requirements, ubiquitous presence of noise and potentially gross corruptions, sometimes with these simultaneously occurring. This paper proposes a Proximal Alternating Robust Subspace Minimization (PARSuMi) method to tackle the three problems. The proximal alternating scheme explicitly exploits the rank constraint on the completed matrix and uses the $\ell_0$ pseudo-norm directly in the corruption recovery step. We show that the proposed method for the non-convex and non-smooth model converges to a stationary point. Although it is not guaranteed to find the global optimal solution, in practice we find that our algorithm can typically arrive at a good local minimizer when it is supplied with a reasonably good starting point based on convex optimization. Extensive experiments with challenging synthetic and real data demonstrate that our algorithm succeeds in a much larger range of practical problems where convex optimization fails, and it also outperforms various state-of-the-art algorithms.