We present a collaborative learning method called Mutual Contrastive Learning (MCL) for general visual representation learning. The core idea of MCL is to perform mutual interaction and transfer of contrastive distributions among a cohort of networks. A crucial component of MCL is Interactive Contrastive Learning (ICL). Compared with vanilla contrastive learning, ICL can aggregate cross-network embedding information and maximize the lower bound to the mutual information between two networks. This enables each network to learn extra contrastive knowledge from others, leading to better feature representations for visual recognition tasks. We emphasize that the resulting MCL is conceptually simple yet empirically powerful. It is a generic framework that can be applied to both supervised and self-supervised representation learning. Experimental results on image classification and transfer learning to object detection show that MCL can lead to consistent performance gains, demonstrating that MCL can guide the network to generate better feature representations. Code is available at https://github.com/winycg/MCL.
The past year has witnessed the rapid development of applying the Transformer module to vision problems. While some researchers have demonstrated that Transformer-based models enjoy a favorable ability of fitting data, there are still growing number of evidences showing that these models suffer over-fitting especially when the training data is limited. This paper offers an empirical study by performing step-by-step operations to gradually transit a Transformer-based model to a convolution-based model. The results we obtain during the transition process deliver useful messages for improving visual recognition. Based on these observations, we propose a new architecture named Visformer, which is abbreviated from the `Vision-friendly Transformer'. With the same computational complexity, Visformer outperforms both the Transformer-based and convolution-based models in terms of ImageNet classification accuracy, and the advantage becomes more significant when the model complexity is lower or the training set is smaller. The code is available at https://github.com/danczs/Visformer.
Convolutional Neural Networks (CNN) have been rigorously studied for Hyperspectral Image Classification (HSIC) and are known to be effective in exploiting joint spatial-spectral information with the expense of lower generalization performance and learning speed due to the hard labels and non-uniform distribution over labels. Several regularization techniques have been used to overcome the aforesaid issues. However, sometimes models learn to predict the samples extremely confidently which is not good from a generalization point of view. Therefore, this paper proposed an idea to enhance the generalization performance of a hybrid CNN for HSIC using soft labels that are a weighted average of the hard labels and uniform distribution over ground labels. The proposed method helps to prevent CNN from becoming over-confident. We empirically show that in improving generalization performance, label smoothing also improves model calibration which significantly improves beam-search. Several publicly available Hyperspectral datasets are used to validate the experimental evaluation which reveals improved generalization performance, statistical significance, and computational complexity as compared to the state-of-the-art models. The code will be made available at https://github.com/mahmad00.
Previous studies have shown the great potential of capsule networks for the spatial contextual feature extraction from {hyperspectral images (HSIs)}. However, the sampling locations of the convolutional kernels of capsules are fixed and cannot be adaptively changed according to the inconsistent semantic information of HSIs. Based on this observation, this paper proposes an adaptive spatial pattern capsule network (ASPCNet) architecture by developing an adaptive spatial pattern (ASP) unit, that can rotate the sampling location of convolutional kernels on the basis of an enlarged receptive field. Note that this unit can learn more discriminative representations of HSIs with fewer parameters. Specifically, two cascaded ASP-based convolution operations (ASPConvs) are applied to input images to learn relatively high-level semantic features, transmitting hierarchical structures among capsules more accurately than the use of the most fundamental features. Furthermore, the semantic features are fed into ASP-based conv-capsule operations (ASPCaps) to explore the shapes of objects among the capsules in an adaptive manner, further exploring the potential of capsule networks. Finally, the class labels of image patches centered on test samples can be determined according to the fully connected capsule layer. Experiments on three public datasets demonstrate that ASPCNet can yield competitive performance with higher accuracies than state-of-the-art methods.
The latest Deep Learning (DL) models for detection and classification have achieved an unprecedented performance over classical machine learning algorithms. However, DL models are black-box methods hard to debug, interpret, and certify. DL alone cannot provide explanations that can be validated by a non technical audience. In contrast, symbolic AI systems that convert concepts into rules or symbols -- such as knowledge graphs -- are easier to explain. However, they present lower generalisation and scaling capabilities. A very important challenge is to fuse DL representations with expert knowledge. One way to address this challenge, as well as the performance-explainability trade-off is by leveraging the best of both streams without obviating domain expert knowledge. We tackle such problem by considering the symbolic knowledge is expressed in form of a domain expert knowledge graph. We present the eXplainable Neural-symbolic learning (X-NeSyL) methodology, designed to learn both symbolic and deep representations, together with an explainability metric to assess the level of alignment of machine and human expert explanations. The ultimate objective is to fuse DL representations with expert domain knowledge during the learning process to serve as a sound basis for explainability. X-NeSyL methodology involves the concrete use of two notions of explanation at inference and training time respectively: 1) EXPLANet: Expert-aligned eXplainable Part-based cLAssifier NETwork Architecture, a compositional CNN that makes use of symbolic representations, and 2) SHAP-Backprop, an explainable AI-informed training procedure that guides the DL process to align with such symbolic representations in form of knowledge graphs. We showcase X-NeSyL methodology using MonuMAI dataset for monument facade image classification, and demonstrate that our approach improves explainability and performance.
Channel Pruning has been long studied to compress CNNs, which significantly reduces the overall computation. Prior works implement channel pruning in an unexplainable manner, which tends to reduce the final classification errors while failing to consider the internal influence of each channel. In this paper, we conduct channel pruning in a white box. Through deep visualization of feature maps activated by different channels, we observe that different channels have a varying contribution to different categories in image classification. Inspired by this, we choose to preserve channels contributing to most categories. Specifically, to model the contribution of each channel to differentiating categories, we develop a class-wise mask for each channel, implemented in a dynamic training manner w.r.t. the input image's category. On the basis of the learned class-wise mask, we perform a global voting mechanism to remove channels with less category discrimination. Lastly, a fine-tuning process is conducted to recover the performance of the pruned model. To our best knowledge, it is the first time that CNN interpretability theory is considered to guide channel pruning. Extensive experiments on representative image classification tasks demonstrate the superiority of our White-Box over many state-of-the-arts. For instance, on CIFAR-10, it reduces 65.23% FLOPs with even 0.62% accuracy improvement for ResNet-110. On ILSVRC-2012, White-Box achieves a 45.6% FLOPs reduction with only a small loss of 0.83% in the top-1 accuracy for ResNet-50.
One of the main challenges since the advancement of convolutional neural networks is how to connect the extracted feature map to the final classification layer. VGG models used two sets of fully connected layers for the classification part of their architectures, which significantly increased the number of models' weights. ResNet and the next deep convolutional models used the Global Average Pooling (GAP) layer to compress the feature map and feed it to the classification layer. Although using the GAP layer reduces the computational cost, but also causes losing spatial resolution of the feature map, which results in decreasing learning efficiency. In this paper, we aim to tackle this problem by replacing the GAP layer with a new architecture called Wise-SrNet. It is inspired by the depthwise convolutional idea and is designed for processing spatial resolution while not increasing computational cost. We have evaluated our method using three different datasets: Intel Image Classification Challenge, MIT Indoors Scenes, and a part of the ImageNet dataset. We investigated the implementation of our architecture on several models of the Inception, ResNet, and DenseNet families. Applying our architecture has revealed a significant effect on increasing convergence speed and accuracy. Our Experiments on images with 224*224 resolution increased the Top-1 accuracy between 2% to 8% on different datasets and models. Running our models on 512*512 resolution images of the MIT Indoors Scenes dataset showed a notable result of improving the Top-1 accuracy within 3% to 26%. We will also demonstrate the GAP layer's disadvantage when the input images are large and the number of classes is not few. In this circumstance, our proposed architecture can do a great help in enhancing classification results. The code is shared at https://github.com/mr7495/image-classification-spatial.
Traditional breast cancer image classification methods require manual extraction of features from medical images, which not only require professional medical knowledge, but also have problems such as time-consuming and labor-intensive and difficulty in extracting high-quality features. Therefore, the paper proposes a computer-based feature fusion Convolutional neural network breast cancer image classification and detection method. The paper pre-trains two convolutional neural networks with different structures, and then uses the convolutional neural network to automatically extract the characteristics of features, fuse the features extracted from the two structures, and finally use the classifier classifies the fused features. The experimental results show that the accuracy of this method in the classification of breast cancer image data sets is 89%, and the classification accuracy of breast cancer images is significantly improved compared with traditional methods.
Random field and random cluster theory are used to describe certain mathematical results concerning the probability distribution of image pixel intensities characterized as generic $2D$ integer arrays. The size of the smallest bounded region within an image is estimated for segmenting an image, from which, the equilibrium distribution of intensities can be recovered. From the estimated bounded regions, properties of the sub-optimal and equilibrium distributions of intensities are derived, which leads to an image compression methodology whereby only slightly more than half of all pixels are required for a worst-case reconstruction of the original image. A custom deep belief network and heuristic allows for the unsupervised segmentation, detection and localization of objects in an image. An example illustrates the mathematical results.
Synthetic aperture radar (SAR) is widely used for ground surface classification since it utilizes information on vegetation and soil unavailable in optical observation. Image classification often employs convolutional neural networks. However, they have serious problems such as long learning time and resolution degradation in their convolution and pooling processes. In this paper, we propose complex-valued reservoir computing (CVRC) to deal with complex-valued images in interferometric SAR (InSAR). We classify InSAR image data by using CVRC successfully with a higher resolution and a lower computational cost, i.e., one-hundredth learning time and one-fifth classification time, than convolutional neural networks. We also conduct experiments on slope angle estimation. CVRC is found applicable to quantitative tasks dealing with continuous values as well as discrete classification tasks with a higher accuracy.
We present Multiscale Vision Transformers (MViT) for video and image recognition, by connecting the seminal idea of multiscale feature hierarchies with transformer models. Multiscale Transformers have several channel-resolution scale stages. Starting from the input resolution and a small channel dimension, the stages hierarchically expand the channel capacity while reducing the spatial resolution. This creates a multiscale pyramid of features with early layers operating at high spatial resolution to model simple low-level visual information, and deeper layers at spatially coarse, but complex, high-dimensional features. We evaluate this fundamental architectural prior for modeling the dense nature of visual signals for a variety of video recognition tasks where it outperforms concurrent vision transformers that rely on large scale external pre-training and are 5-10x more costly in computation and parameters. We further remove the temporal dimension and apply our model for image classification where it outperforms prior work on vision transformers. Code is available at: https://github.com/facebookresearch/SlowFast
We present a framework for learning multimodal representations from unlabeled data using convolution-free Transformer architectures. Specifically, our Video-Audio-Text Transformer (VATT) takes raw signals as inputs and extracts multimodal representations that are rich enough to benefit a variety of downstream tasks. We train VATT end-to-end from scratch using multimodal contrastive losses and evaluate its performance by the downstream tasks of video action recognition, audio event classification, image classification, and text-to-video retrieval. Furthermore, we study a modality-agnostic, single-backbone Transformer by sharing weights among the three modalities. We show that the convolution-free VATT outperforms state-of-the-art ConvNet-based architectures in the downstream tasks. Especially, VATT's vision Transformer achieves the top-1 accuracy of 82.1% on Kinetics-400, 83.6% on Kinetics-600, 72.7% on Kinetics-700, and 41.1% on Moments in Time, new records while avoiding supervised pre-training. Transferring to image classification leads to 78.7% top-1 accuracy on ImageNet compared to 64.7% by training the same Transformer from scratch, showing the generalizability of our model despite the domain gap between videos and images. VATT's audio Transformer also sets a new record on waveform-based audio event recognition by achieving the mAP of 39.4% on AudioSet without any supervised pre-training. VATT's source code is publicly available.
ImageNet-1K serves as the primary dataset for pretraining deep learning models for computer vision tasks. ImageNet-21K dataset, which is bigger and more diverse, is used less frequently for pretraining, mainly due to its complexity, low accessibility, and underestimation of its added value. This paper aims to close this gap, and make high-quality efficient pretraining on ImageNet-21K available for everyone. Via a dedicated preprocessing stage, utilization of WordNet hierarchical structure, and a novel training scheme called semantic softmax, we show that various models significantly benefit from ImageNet-21K pretraining on numerous datasets and tasks, including small mobile-oriented models. We also show that we outperform previous ImageNet-21K pretraining schemes for prominent new models like ViT and Mixer. Our proposed pretraining pipeline is efficient, accessible, and leads to SoTA reproducible results, from a publicly available dataset. The training code and pretrained models are available at: https://github.com/Alibaba-MIIL/ImageNet21K
In this paper, we present token labeling -- a new training objective for training high-performance vision transformers (ViTs). Different from the standard training objective of ViTs that computes the classification loss on an additional trainable class token, our proposed one takes advantage of all the image patch tokens to compute the training loss in a dense manner. Specifically, token labeling reformulates the image classification problem into multiple token-level recognition problems and assigns each patch token with an individual location-specific supervision generated by a machine annotator. Experiments show that token labeling can clearly and consistently improve the performance of various ViT models across a wide spectrum. For a vision transformer with 26M learnable parameters serving as an example, with token labeling, the model can achieve 84.4% Top-1 accuracy on ImageNet. The result can be further increased to 86.4% by slightly scaling the model size up to 150M, delivering the minimal-sized model among previous models (250M+) reaching 86%. We also show that token labeling can clearly improve the generalization capability of the pre-trained models on downstream tasks with dense prediction, such as semantic segmentation. Our code and all the training details will be made publicly available at https://github.com/zihangJiang/TokenLabeling.
Optimized for pixel fidelity metrics, images compressed by existing image codec are facing systematic challenges when used for visual analysis tasks, especially under low-bitrate coding. This paper proposes a visual analysis-motivated rate-distortion model for Versatile Video Coding (VVC) intra compression. The proposed model has two major contributions, a novel rate allocation strategy and a new distortion measurement model. We first propose the region of interest for machine (ROIM) to evaluate the degree of importance for each coding tree unit (CTU) in visual analysis. Then, a novel CTU-level bit allocation model is proposed based on ROIM and the local texture characteristics of each CTU. After an in-depth analysis of multiple distortion models, a visual analysis friendly distortion criteria is subsequently proposed by extracting deep feature of each coding unit (CU). To alleviate the problem of lacking spatial context information when calculating the distortion of each CU, we finally propose a multi-scale feature distortion (MSFD) metric using different neighboring pixels by weighting the extracted deep features in each scale. Extensive experimental results show that the proposed scheme could achieve up to 28.17\% bitrate saving under the same analysis performance among several typical visual analysis tasks such as image classification, object detection, and semantic segmentation.
We study the problem of directly optimizing arbitrary non-differentiable task evaluation metrics such as misclassification rate and recall. Our method, named MetricOpt, operates in a black-box setting where the computational details of the target metric are unknown. We achieve this by learning a differentiable value function, which maps compact task-specific model parameters to metric observations. The learned value function is easily pluggable into existing optimizers like SGD and Adam, and is effective for rapidly finetuning a pre-trained model. This leads to consistent improvements since the value function provides effective metric supervision during finetuning, and helps to correct the potential bias of loss-only supervision. MetricOpt achieves state-of-the-art performance on a variety of metrics for (image) classification, image retrieval and object detection. Solid benefits are found over competing methods, which often involve complex loss design or adaptation. MetricOpt also generalizes well to new tasks and model architectures.
Deep learning typically requires large data sets and much compute power for each new problem that is learned. Meta-learning can be used to learn a good prior that facilitates quick learning, thereby relaxing these requirements so that new tasks can be learned quicker; two popular approaches are MAML and the meta-learner LSTM. In this work, we compare the two and formally show that the meta-learner LSTM subsumes MAML. Combining this insight with recent empirical findings, we construct a new algorithm (dubbed TURTLE) which is simpler than the meta-learner LSTM yet more expressive than MAML. TURTLE outperforms both techniques at few-shot sine wave regression and image classification on miniImageNet and CUB without any additional hyperparameter tuning, at a computational cost that is comparable with second-order MAML. The key to TURTLE's success lies in the use of second-order gradients, which also significantly increases the performance of the meta-learner LSTM by 1-6% accuracy.
Machine learning systems typically assume that the distributions of training and test sets match closely. However, a critical requirement of such systems in the real world is their ability to generalize to unseen domains. Here, we propose an inter-domain gradient matching objective that targets domain generalization by maximizing the inner product between gradients from different domains. Since direct optimization of the gradient inner product can be computationally prohibitive -- requires computation of second-order derivatives -- we derive a simpler first-order algorithm named Fish that approximates its optimization. We demonstrate the efficacy of Fish on 6 datasets from the Wilds benchmark, which captures distribution shift across a diverse range of modalities. Our method produces competitive results on these datasets and surpasses all baselines on 4 of them. We perform experiments on both the Wilds benchmark, which captures distribution shift in the real world, as well as datasets in DomainBed benchmark that focuses more on synthetic-to-real transfer. Our method produces competitive results on both benchmarks, demonstrating its effectiveness across a wide range of domain generalization tasks.
We propose a framework using contrastive learning as a pre-training task to perform image classification in the presence of noisy labels. Recent strategies such as pseudo-labeling, sample selection with Gaussian Mixture models, weighted supervised contrastive learning have been combined into a fine-tuning phase following the pre-training. This paper provides an extensive empirical study showing that a preliminary contrastive learning step brings a significant gain in performance when using different loss functions: non-robust, robust, and early-learning regularized. Our experiments performed on standard benchmarks and real-world datasets demonstrate that: i) the contrastive pre-training increases the robustness of any loss function to noisy labels and ii) the additional fine-tuning phase can further improve accuracy but at the cost of additional complexity.
Machine learning with deep neural networks (DNNs) has become one of the foundation techniques in many safety-critical systems, such as autonomous vehicles and medical diagnosis systems. DNN-based systems, however, are known to be vulnerable to adversarial examples (AEs) that are maliciously perturbed variants of legitimate inputs. While there has been a vast body of research to defend against AE attacks in the literature, the performances of existing defense techniques are still far from satisfactory, especially for adaptive attacks, wherein attackers are knowledgeable about the defense mechanisms and craft AEs accordingly. In this work, we propose a multilayer defense-in-depth framework for AE detection, namely MixDefense. For the first layer, we focus on those AEs with large perturbations. We propose to leverage the `noise' features extracted from the inputs to discover the statistical difference between natural images and tampered ones for AE detection. For AEs with small perturbations, the inference result of such inputs would largely deviate from their semantic information. Consequently, we propose a novel learning-based solution to model such contradictions for AE detection. Both layers are resilient to adaptive attacks because there do not exist gradient propagation paths for AE generation. Experimental results with various AE attack methods on image classification datasets show that the proposed MixDefense solution outperforms the existing AE detection techniques by a considerable margin.