Being data-driven is one of the most iconic properties of deep learning algorithms. The birth
of ImageNet drives a remarkable trend of" learning from large-scale data" in computer vision …

Multi-view projection methods have demonstrated their ability to reach state-of-the-art
performance on 3D shape recognition. Those methods learn different ways to aggregate …

Deep neural networks are vulnerable to adversarial attacks, in which imperceptible
perturbations to their input lead to erroneous network predictions. This phenomenon has …

Distribution shift is a major source of failure for machine learning models. However,
evaluating model reliability under distribution shift can be challenging, especially since it …

Visual recognition models are not invariant to viewpoint changes in the 3D world, as
different viewing directions can dramatically affect the predictions given the same object …

Verifying robustness of neural networks given a specified threat model is a fundamental yet
challenging task. While current verification methods mainly focus on the l_p-norm threat …

We introduce 3DB: an extendable, unified framework for testing and debugging vision
models using photorealistic simulation. We demonstrate, through a wide range of use cases …

Viewpoint invariance remains challenging for visual recognition in the 3D world, as altering
the viewing directions can significantly impact predictions for the same object. While …

We introduce DeepCert, a tool-supported method for verifying the robustness of deep neural
network (DNN) image classifiers to contextually relevant perturbations such as blur, haze …

One major factor impeding more widespread adoption of deep neural networks (DNNs) is
their lack of robustness, which is essential for safety-critical applications such as …