Recently, there has been an increasing interest in applying attention mechanisms in Convolutional Neural Networks (CNNs) to solve computer vision tasks. Most of these methods learn to explicitly identify and highlight relevant parts of the scene and pass the attended image to further layers of the network. In this paper, we argue that such an approach might not be optimal. Arguably, explicitly learning which parts of the image are relevant is typically harder than learning which parts of the image are less relevant and, thus, should be ignored. In fact, in vision domain, there are many easy-to-identify patterns of irrelevant features. For example, image regions close to the borders are less likely to contain useful information for a classification task. Based on this idea, we propose to reformulate the attention mechanism in CNNs to learn to ignore instead of learning to attend. Specifically, we propose to explicitly learn irrelevant information in the scene and suppress it in the produced representation, keeping only important attributes. This implicit attention scheme can be incorporated into any existing attention mechanism. In this work, we validate this idea using two recent attention methods Squeeze and Excitation (SE) block and Convolutional Block Attention Module (CBAM). Experimental results on different datasets and model architectures show that learning to ignore, i.e., implicit attention, yields superior performance compared to the standard approaches.
Most scenes are illuminated by several light sources, where the traditional assumption of uniform illumination is invalid. This issue is ignored in most color constancy methods, primarily due to the complex spatial impact of multiple light sources on the image. Moreover, most existing multi-illuminant methods fail to preserve the smooth change of illumination, which stems from spatial dependencies in natural images. Motivated by this, we propose a novel multi-illuminant color constancy method, by learning pixel-wise illumination maps caused by multiple light sources. The proposed method enforces smoothness within neighboring pixels, by regularizing the training with the total variation loss. Moreover, a bilateral filter is provisioned further to enhance the natural appearance of the estimated images, while preserving the edges. Additionally, we propose a label-smoothing technique that enables the model to generalize well despite the uncertainties in ground truth. Quantitative and qualitative experiments demonstrate that the proposed method outperforms the state-of-the-art.
In this paper, we consider the problem of non-linear dimensionality reduction under uncertainty, both from a theoretical and algorithmic perspectives. Since real-world data usually contain measurements with uncertainties and artifacts, the input space in the proposed framework consists of probability distributions to model the uncertainties associated with each sample. We propose a new dimensionality reduction framework, called NGEU, which leverages uncertainty information and directly extends several traditional approaches, e.g., KPCA, MDA/KMFA, to receive as inputs the probability distributions instead of the original data. We show that the proposed NGEU formulation exhibits a global closed-form solution, and we analyze, based on the Rademacher complexity, how the underlying uncertainties theoretically affect the generalization ability of the framework. Empirical results on different datasets show the effectiveness of the proposed framework.
Management of unstructured production data is a key challenge for Industry 4.0. Effective product tracking endorses data integration and productivity improvements throughout the manufacturing processes. Radio-frequency identification (RFID) tags are used in many tracking cases, but in some manufacturing environments, those cannot be used as they might get damaged or removed during processing. In this paper, we propose an alternative visual product tracking system. The physical system uses two cameras placed at the two ends of the tracked process(es). Product pairs are then matched with a Siamese neural network operating on the product images and trained offline on the problem at hand with labeled data. The proposed system can track products solely based on their visual appearance and without any physical interference with the products or production processes. Unlike other existing image-based methods, the proposed system is invariant to major positional and visual changes in the products. As a proof-of-concept, we tested the proposed system with real plywood factory data and were able to track the products with 98.5 % accuracy in a realistic test scenario. The implementation of the proposed method and the Veneer21 dataset are publicly available at https://github.com/TuomasJalonen/visual-product-tracking-system.
Neural networks are composed of multiple layers arranged in a hierarchical structure jointly trained with a gradient-based optimization, where the errors are back-propagated from the last layer back to the first one. At each optimization step, neurons at a given layer receive feedback from neurons belonging to higher layers of the hierarchy. In this paper, we propose to complement this traditional 'between-layer' feedback with additional 'within-layer' feedback to encourage the diversity of the activations within the same layer. To this end, we measure the pairwise similarity between the outputs of the neurons and use it to model the layer's overall diversity. We present an extensive empirical study confirming that the proposed approach enhances the performance of several state-of-the-art neural network models in multiple tasks. The code is publically available at \url{https://github.com/firasl/AAAI-23-WLD-Reg}
In this paper, we describe a new large dataset for illumination estimation. This dataset, called INTEL-TAU, contains 7022 images in total, which makes it the largest available high-resolution dataset for illumination estimation research. The variety of scenes captured using three different camera models, namely Canon 5DSR, Nikon D810, and Sony IMX135, makes the dataset appropriate for evaluating the camera and scene invariance of the different illumination estimation techniques. Privacy masking is done for sensitive information, e.g., faces. Thus, the dataset is coherent with the new General Data Protection Regulation (GDPR). Furthermore, the effect of color shading for mobile images can be evaluated with INTEL-TAU dataset, as both corrected and uncorrected versions of the raw data are provided. Furthermore, this paper benchmarks several color constancy approaches on the proposed dataset.
Recently, there has been an increasing interest in applying attention mechanisms in Convolutional Neural Networks (CNNs) to solve computer vision tasks. Most of these methods learn to explicitly identify and highlight relevant parts of the scene and pass the attended image to further layers of the network. In this paper, we argue that such an approach might not be optimal. Arguably, explicitly learning which parts of the image are relevant is typically harder than learning which parts of the image are less relevant and, thus, should be ignored. In fact, in vision domain, there are many easy-to-identify patterns of irrelevant features. For example, image regions close to the borders are less likely to contain useful information for a classification task. Based on this idea, we propose to reformulate the attention mechanism in CNNs to learn to ignore instead of learning to attend. Specifically, we propose to explicitly learn irrelevant information in the scene and suppress it in the produced representation, keeping only important attributes. This implicit attention scheme can be incorporated into any existing attention mechanism. In this work, we validate this idea using two recent attention methods Squeeze and Excitation (SE) block and Convolutional Block Attention Module (CBAM). Experimental results on different datasets and model architectures show that learning to ignore, i.e., implicit attention, yields superior performance compared to the standard approaches.
Autoencoders are a type of unsupervised neural networks, which can be used to solve various tasks, e.g., dimensionality reduction, image compression, and image denoising. An AE has two goals: (i) compress the original input to a low-dimensional space at the bottleneck of the network topology using an encoder, (ii) reconstruct the input from the representation at the bottleneck using a decoder. Both encoder and decoder are optimized jointly by minimizing a distortion-based loss which implicitly forces the model to keep only those variations of input data that are required to reconstruct the and to reduce redundancies. In this paper, we propose a scheme to explicitly penalize feature redundancies in the bottleneck representation. To this end, we propose an additional loss term, based on the pair-wise correlation of the neurons, which complements the standard reconstruction loss forcing the encoder to learn a more diverse and richer representation of the input. We tested our approach across different tasks: dimensionality reduction using three different dataset, image compression using the MNIST dataset, and image denoising using fashion MNIST. The experimental results show that the proposed loss leads consistently to superior performance compared to the standard AE loss.
Recently, Convolutional Neural Networks (CNNs) have been widely used to solve the illuminant estimation problem and have often led to state-of-the-art results. Standard approaches operate directly on the input image. In this paper, we argue that this problem can be decomposed into three channel-wise independent and symmetric sub-problems and propose a novel CNN-based illumination estimation approach based on this decomposition. The proposed method substantially reduces the number of parameters needed to solve the task while achieving competitive experimental results compared to state-of-the-art methods. Furthermore, the practical application of illumination estimation techniques typically requires identifying the extreme error cases. This can be achieved using an uncertainty estimation technique. In this work, we propose a novel color constancy uncertainty estimation approach that augments the trained model with an auxiliary branch which learns to predict the error based on the feature representation. Intuitively, the model learns which feature combinations are robust and are thus likely to yield low errors and which combinations result in erroneous estimates. We test this approach on the proposed method and show that it can indeed be used to avoid several extreme error cases and, thus, improves the practicality of the proposed technique.
