Search for: All records

Using offline training schemes, researchers have tackled the event segmentation problem by providing full or weak-supervision through manually annotated labels or self-supervised epoch-based training. Most works consider videos that are at most 10’s of minutes long. We present a self-supervised perceptual prediction framework capable of temporal event segmentation by building stable representations of objects over time and demonstrate it on long videos, spanning several days at 25 FPS. The approach is deceptively simple but quite effective. We rely on predictions of high-level features computed by a standard deep learning backbone. For prediction, we use an LSTM, augmented with an attention mechanism, trained in a self-supervised manner using the prediction error. The self-learned attention maps effectively localize and track the event-related objects in each frame. The proposed approach does not require labels. It requires only a single pass through the video, with no separate training set. Given the lack of datasets of very long videos, we demonstrate our method on video from 10 d (254 h) of continuous wildlife monitoring data that we had collected with required permissions. We find that the approach is robust to various environmental conditions such as day/night conditions, rain, sharp shadows, and windy conditions. For the task of temporally locating events at the activity level, we had an 80% activity recall rate for one false activity detection every 50 min. We will make the dataset, which is the first of its kind, and the code available to the research community. Project page is available at https://ramymounir.com/publications/EventSegmentation/. 

Fingerprint-based authentication has been successfully adopted in a wide range of applications, including law enforcement and immigration, due to its numerous advantages over traditional password-based authentication. However, despite the usability and accuracy of this technology, some significant concerns still exist, which can potentially hinder its further adoption. For instance, a subject’s fingerprint is permanently associated with an individual and, once stolen, cannot be replaced, thus compromising biometric-based authentication. To mitigate this concern, we propose a multi-factor authentication approach that integrates type 1 and type 3 authentication factors into a fingerprint-based personal identification number, or FingerPIN. To authenticate, a subject is required to present a sequence of fingerprints corresponding to the digits of the PIN, based on a predefined secret mapping between digits and fingers. We conduct a vulnerability analysis of the proposed scheme, and demonstrate that it is robust to the compromise of one or more of the subject’s fingerprints.