Abstract: We propose a new method for exemplar-free class incremental training of ViTs. The main challenge of exemplar-free continual learning is maintaining plasticity of the learner without causing catastrophic forgetting of previously learned tasks. This is often achieved via exemplar replay which can help recalibrate previous task classifiers to the feature drift which occurs when learning new tasks. Exemplar replay, however, comes at the cost of retaining samples from previous tasks which for many applications may not be possible. To address the problem of continual ViT training, we first propose gated class-attention to minimize the drift in the final ViT transformer block. This mask-based gating is applied to class-attention mechanism of the last transformer block and strongly regulates the weights crucial for previous tasks. Importantly, gated class-attention does not require the task-ID during inference, which distinguishes it from other parameter isolation methods. Secondly, we propose a new method of feature drift compensation that accommodates feature drift in the backbone when learning new tasks. The combination of gated class-attention and cascaded feature drift compensation allows for plasticity towards new tasks while limiting forgetting of previous ones. Extensive experiments performed on CIFAR-100, Tiny-ImageNet and ImageNet100 demonstrate that our exemplar-free method obtains competitive results when compared to rehearsal based ViT methods.

Abstract: Recent advances in camera designs and imaging pipelines allow us to capture high-quality images using smartphones. However, due to the small size and lens limitations of the smartphone cameras, we commonly find artifacts or degradation in the processed images. The most common unpleasant effects are noise artifacts, diffraction artifacts, blur, and HDR overexposure. Deep learning methods for image restoration can successfully remove these artifacts. However, most approaches are not suitable for real-time applications on mobile devices due to their heavy computation and memory requirements. In this paper, we propose LPIENet, a lightweight network for perceptual image enhancement, with the focus on deploying it on smartphones. Our experiments show that, with much fewer parameters and operations, our model can deal with the mentioned artifacts and achieve competitive performance compared with state-of-the-art methods on standard benchmarks. Moreover, to prove the efficiency and reliability of our approach, we deployed the model directly on commercial smartphones and evaluated its performance. Our model can process 2K resolution images under 1 second in mid-level commercial smartphones.

Abstract: In the film industry, the same movie is expected to be watched on displays of vastly different sizes, from cinema screens to mobile phones. But visual induction, the perceptual phenomenon by which the appearance of a scene region is affected by its surroundings, will be different for the same image shown on two displays of different dimensions. This phenomenon presents a practical challenge for the preservation of the artistic intentions of filmmakers, because it can lead to shifts in image appearance between viewing destinations. In this work, we show that a neural field model based on the efficient representation principle is able to predict induction effects and how, by regularizing its associated energy functional, the model is still able to represent induction but is now invertible. From this finding, we propose a method to preprocess an image in a screen–size dependent way so that its perception, in terms of visual induction, may remain constant across displays of different size. The potential of the method is demonstrated through psychophysical experiments on synthetic images and qualitative examples on natural images.

Abstract: When extracting information from handwritten documents, text transcription and named entity recognition are usually faced as separate subsequent tasks. This has the disadvantage that errors in the first module affect heavily the
performance of the second module. In this work we propose to do both tasks jointly, using a single neural network with a common architecture used for plain text recognition. Experimentally, the work has been tested on a collection of historical marriage records. Results of experiments are presented to show the effect on the performance for different
configurations: different ways of encoding the information, doing or not transfer learning and processing at text line or multi-line region level. The results are comparable to state of the art reported in the ICDAR 2017 Information Extraction competition, even though the proposed technique does not use any dictionaries, language modeling or post processing.

Abstract: Research on continual learning has led to a variety of approaches to mitigating catastrophic forgetting in feed-forward classification networks. Until now surprisingly little attention has been focused on continual learning of recurrent models applied to problems like image captioning. In this paper we take a systematic look at continual learning of LSTM-based models for image captioning. We propose an attention-based approach that explicitly accommodates the transient nature of vocabularies in continual image captioning tasks -- i.e. that task vocabularies are not disjoint. We call our method Recurrent Attention to Transient Tasks (RATT), and also show how to adapt continual learning approaches based on weight egularization and knowledge distillation to recurrent continual learning problems. We apply our approaches to incremental image captioning problem on two new continual learning benchmarks we define using the MS-COCO and Flickr30 datasets. Our results demonstrate that RATT is able to sequentially learn five captioning tasks while incurring no forgetting of previously learned ones.

Abstract: This paper presents a multi-view scheme to tackle the challenging problem of the self-occlusion in human pose estimation problem. The proposed approach first obtains the human body joints of a set of images, which are captured from different views at the same time. Then, it enhances the obtained joints by using a
multi-view scheme. Basically, the joints from a given view are used to enhance poorly estimated joints from another view, especially intended to tackle the self occlusions cases. A network architecture initially proposed for the monocular case is adapted to be used in the proposed multi-view scheme. Experimental results and
comparisons with the state-of-the-art approaches on Human3.6m dataset are presented showing improvements in the accuracy of body joints estimations.