Abstract: Transformer models have shown great success handling long-range interactions, making them a promising tool for modeling video. However, they lack inductive biases and scale quadratically with input length. These limitations are further exacerbated when dealing with the high dimensionality introduced by the temporal dimension. While there are surveys analyzing the advances of Transformers for vision, none focus on an in-depth analysis of video-specific designs. In this survey, we analyze the main contributions and trends of works leveraging Transformers to model video. Specifically, we delve into how videos are handled at the input level first. Then, we study the architectural changes made to deal with video more efficiently, reduce redundancy, re-introduce useful inductive biases, and capture long-term temporal dynamics. In addition, we provide an overview of different training regimes and explore effective self-supervised learning strategies for video. Finally, we conduct a performance comparison on the most common benchmark for Video Transformers (i.e., action classification), finding them to outperform 3D ConvNets even with less computational complexity.

Abstract: In this work we present the iterative multi-class multi-scale stacked sequential learning framework (IMMSSL), a novel learning scheme that is particularly suited for medical volume segmentation applications. This model exploits the inherent voxel contextual information of the structures of interest in order to improve its segmentation performance results. Without any feature set or learning algorithm prior assumption, the proposed scheme directly seeks to learn the contextual properties of a region from the predicted classifications of previous classifiers within an iterative scheme. Performance results regarding segmentation accuracy in three two-class and multi-class medical volume datasets show a significant improvement with respect to state of the art alternatives. Due to its easiness of implementation and its independence of feature space and learning algorithm, the presented machine learning framework could be taken into consideration as a first choice in complex volume segmentation scenarios.

Abstract: PR4883, PII: S0031-3203(13)00327-0
Two-dimensional shape models have been successfully applied to solve many problems in computer vision, such as object tracking, recognition, and segmentation. Typically, 2D shape models are learned from a discrete set of image landmarks (corresponding to projection of 3D points of an object), after applying Generalized Procustes Analysis (GPA) to remove 2D rigid transformations. However, the
standard GPA process suffers from three main limitations. Firstly, the 2D training samples do not necessarily cover a uniform sampling of all the 3D transformations of an object. This can bias the estimate of the shape model. Secondly, it can be computationally expensive to learn the shape model by sampling 3D transformations. Thirdly, standard GPA methods use only one reference shape, which can might be insufficient to capture large structural variability of some objects.
To address these drawbacks, this paper proposes continuous generalized Procrustes analysis (CGPA).
CGPA uses a continuous formulation that avoids the need to generate 2D projections from all the rigid 3D transformations. It builds an efficient (in space and time) non-biased 2D shape model from a set of 3D model of objects. A major challenge in CGPA is the need to integrate over the space of 3D rotations, especially when the rotations are parameterized with Euler angles. To address this problem, we introduce the use of the Haar measure. Finally, we extended CGPA to incorporate several reference shapes. Experimental results on synthetic and real experiments show the benefits of CGPA over GPA.