Abstract: Multimodal representations and continual learning are two areas closely related to human intelligence. The former considers the learning of shared representation spaces where information from different modalities can be compared and integrated (we focus on cross-modal retrieval between language and visual representations). The latter studies how to prevent forgetting a previously learned task when learning a new one. While humans excel in these two aspects, deep neural networks are still quite limited. In this paper, we propose a combination of both problems into a continual cross-modal retrieval setting, where we study how the catastrophic interference caused by new tasks impacts the embedding spaces and their cross-modal alignment required for effective retrieval. We propose a general framework that decouples the training, indexing and querying stages. We also identify and study different factors that may lead to forgetting, and propose tools to alleviate it. We found that the indexing stage pays an important role and that simply avoiding reindexing the database with updated embedding networks can lead to significant gains. We evaluated our methods in two image-text retrieval datasets, obtaining significant gains with respect to the fine tuning baseline.

Abstract: Online continual learning aims to learn from a non-IID stream of data from a number of different tasks, where the learner is only allowed to consider data once. Methods are typically allowed to use a limited buffer to store some of the images in the stream. Recently, it was found that feature replay, where an intermediate layer representation of the image is stored (or generated) leads to superior results than image replay, while requiring less memory. Quantized exemplars can further reduce the memory usage. However, a drawback of these methods is that they use a fixed (or very intransigent) backbone network. This significantly limits the learning of representations that can discriminate between all tasks. To address this problem, we propose an auxiliary classifier auto-encoder (ACAE) module for feature replay at intermediate layers with high compression rates. The reduced memory footprint per image allows us to save more exemplars for replay. In our experiments, we conduct task-agnostic evaluation under online continual learning setting and get state-of-the-art performance on ImageNet-Subset, CIFAR100 and CIFAR10 dataset.

Abstract: Self-supervised learning has been widely applied to train high-quality vision transformers. Unleashing their excellent performance on memory and compute constraint devices is therefore an important research topic. However, how to distill knowledge from one self-supervised ViT to another has not yet been explored. Moreover, the existing self-supervised knowledge distillation (SSKD) methods focus on ConvNet based architectures are suboptimal for ViT knowledge distillation. In this paper, we study knowledge distillation of self-supervised vision transformers (ViT-SSKD). We show that directly distilling information from the crucial attention mechanism from teacher to student can significantly narrow the performance gap between both. In experiments on ImageNet-Subset and ImageNet-1K, we show that our method AttnDistill outperforms existing self-supervised knowledge distillation (SSKD) methods and achieves state-of-the-art k-NN accuracy compared with self-supervised learning (SSL) methods learning from scratch (with the ViT-S model). We are also the first to apply the tiny ViT-T model on self-supervised learning. Moreover, AttnDistill is independent of self-supervised learning algorithms, it can be adapted to ViT based SSL methods to improve the performance in future research.

Abstract: Deep learning has drastically changed computer vision in the past decades and achieved great success in many applications, such as image classification, retrieval, detection, and segmentation thanks to the emergence of neural networks. Typically, for most applications, these networks are presented with examples from all tasks they are expected to perform. However, for many applications, this is not a realistic
scenario, and an algorithm is required to learn tasks sequentially. Continual learning proposes theory and methods for this scenario.
The main challenge for continual learning systems is called catastrophic forgetting and refers to a significant drop in performance on previous tasks. To tackle this problem, three main branches of methods have been explored to alleviate the forgetting in continual learning. They are regularization-based methods, rehearsalbased methods, and parameter isolation-based methods. However, most of them are focused on image classification tasks. Continual learning of many computer vision fields has still not been well-explored. Thus, in this thesis, we extend the continual learning knowledge to meta learning, we propose a method for the incremental learning of hierarchical relations for image classification, we explore image recognition in online continual learning, and study continual learning for cross-modal learning.
In this thesis, we explore the usage of image rehearsal when addressing the incremental meta learning problem. Observing that existingmethods fail to improve performance with saved exemplars, we propose to mix exemplars with current task data and episode-level distillation to overcome forgetting in incremental meta learning. Next, we study a more realistic image classification scenario where each class has multiple granularity levels. Only one label is present at any time, which requires the model to infer if the provided label has a hierarchical relation with any already known label. In experiments, we show that the estimated hierarchy information can be beneficial in both the training and inference stage.
For the online continual learning setting, we investigate the usage of intermediate feature replay. In this case, the training samples are only observed by the model only one time. Here we fix thememory buffer for feature replay and compare the effectiveness of saving features from different layers. Finally, we investigate multi-modal continual learning, where an image encoder is cooperating with a semantic branch. We consider the continual learning of both zero-shot learning and cross-modal retrieval problems.

Abstract: Class incremental semantic segmentation (CISS) focuses on alleviating catastrophic forgetting to improve discrimination. Previous work mainly exploit regularization (e.g., knowledge distillation) to maintain previous knowledge in the current model. However, distillation alone often yields limited gain to the model since only the representations of old and new models are restricted to be consistent. In this paper, we propose a simple yet effective method to obtain a model with strong memory of old knowledge, named Endpoints Weight Fusion (EWF). In our method, the model containing old knowledge is fused with the model retaining new knowledge in a dynamic fusion manner, strengthening the memory of old classes in ever-changing distributions. In addition, we analyze the relation between our fusion strategy and a popular moving average technique EMA, which reveals why our method is more suitable for class-incremental learning. To facilitate parameter fusion with closer distance in the parameter space, we use distillation to enhance the optimization process. Furthermore, we conduct experiments on two widely used datasets, achieving the state-of-the-art performance.

Abstract: In this paper, we present two large video multi-modal datasets for RGB and RGB-D gesture recognition: the ChaLearn LAP RGB-D Isolated Gesture Dataset (IsoGD)and the Continuous Gesture Dataset (ConGD). Both datasets are derived from the ChaLearn Gesture Dataset
(CGD) that has a total of more than 50000 gestures for the “one-shot-learning” competition. To increase the potential of the old dataset, we designed new well curated datasets composed of 249 gesture labels, and including 47933 gestures manually labeled the begin and end frames in sequences.Using these datasets we will open two competitions
on the CodaLab platform so that researchers can test and compare their methods for “user independent” gesture recognition. The first challenge is designed for gesture spotting
and recognition in continuous sequences of gestures while the second one is designed for gesture classification from segmented data. The baseline method based on the bag of visual words model is also presented.

Abstract: This guest editorial introduces the twenty two papers accepted for this Special Issue on Articulated Motion and Deformable Objects (AMDO). They are grouped into four main categories within the field of AMDO: human motion analysis (action/gesture), human pose estimation, deformable shape segmentation, and face analysis. For each of the four topics, a survey of the recent developments in the field is presented. The accepted papers are briefly introduced in the context of this survey. They contribute novel methods, algorithms with improved performance as measured on benchmarking datasets, as well as two new datasets for hand action detection and human posture analysis. The special issue should be of high relevance to the reader interested in AMDO recognition and promote future research directions in the field.