Neural Information Processing Systems http://nips.cc/

The state-of-the-art deep neural networks (DNNs) are vulnerable to adversarial examples with additive random noise-like perturbations. While such examples are hardly found in the physical world, the image blurring effect caused by object motion, on the other hand, commonly occurs in practice, making the study of which greatly important especially for the widely adopted real-time image processing tasks ( e.g ., object detection, tracking). In this paper, we initiate the first step to comprehensively investigate the potential hazards of blur effect for DNN, caused by object motion. We propose a novel adversarial attack method that can generate visually natural motion-blurred adversarial examples, named motion-based adversarial blur attack (AB BA).

Work performed while at MIT. 34th Conference on Neural Information Processing Systems (NeurIPS 2020), V ancouver, Canada. Under Assumption 1, P (M) is a cubic ( i.e.,, 3-regular), planar graph. The medial graph of a cubic plane graph coincides with its line graph. The proposition follows directly from Lemmas 1 and 2.Lemma 3 In the following, we show that PD-MeshNet can be easily extended to handle non-manifold meshes. Figure C.1: The edge collapse operation used for pooling in [ We provide in the following a detail about the definition of the dual features.

We propose a method that combines the advantages of both types of approaches, while addressing their limitations: we extend a primal-dual framework drawn from the graph-neural-network literature to triangle meshes, and define convolutions on two types of graphs constructed from an input mesh. Our method takes features for both edges and faces of a 3D mesh as input, and dynamically aggregates them using an attention mechanism.

Instead of considering "online" and "continual" learning as inconvenient constraints to avoid, in

Neural Information Processing Systems http://nips.cc/

Neural Information Processing Systems http://nips.cc/

Hence, R5 is wary of the relevance of our examples to deep learning.

This heavily restricts the degree of control over generated objects as well as the possible applications of such models.