In this demonstration, we will present EVA, an end-to-end AI-Relational database management system. We will demonstrate the capabilities and utility of EVA using three usage scenarios: (1) EVA serves as a backend for an exploratory video analytics interface developed using Streamlit and React, (2) EVA seamlessly integrates with the Python and Data Science ecosystems by allowing users to access EVA in a Python notebook alongside other popular libraries such as Pandas and Matplotlib, and (3) EVA facilitates bulk labeling with Label Studio, a widely-used labeling framework. By optimizing complex vision queries, we illustrate how EVA allows a wide range of application developers to harness the recent advances in computer vision.
more » « less- PAR ID:
- 10483746
- Publisher / Repository:
- VLDB
- Date Published:
- Journal Name:
- Proceedings of the VLDB Endowment
- Volume:
- 16
- Issue:
- 12
- ISSN:
- 2150-8097
- Page Range / eLocation ID:
- 4082 to 4085
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
Event and frame cameras capture the complemen-tary spatial and temporal details of a scene providing an accuracy vs. latency trade-off. Fusing these processing modalities using convolutional (CNN) and spiking neural networks (SNN) respectively has been shown for target tracking. We present our heterogeneous RRAM compute-in-memory (CIM) and SRAM compute-near-memory (CNM) SoC for simultaneous processing of CNN and SNN. We will show the advantage of using fused vision over frame-only vision and demonstrate python programmable data streaming. The visitors will be able to see the processing-dependent dynamic power gating of non-volatile RRAM and in-memory error correction capability.more » « less
-
In recent years, deep learning models have revolutionized computer vision, enabling diverse applications. However, these models are computationally expensive, and leveraging them for video analyt- ics involves low-level imperative programming. To address these efficiency and usability challenges, the database community has de- veloped video database management systems (VDBMSs). However, existing VDBMSs lack extensibility and composability and do not support holistic system optimizations, limiting their practical appli- cation. In response to these issues, we present our vision for EVA, a VDBMS that allows for extensible support of user-defined functions and employs a Cascades-style query optimizer. Additionally, we leverage RAY’s distributed execution to enhance scalability and performance and explore hardware-specific optimizations to facilitate runtime optimizations. We discuss the architecture and design of EVA, our achievements thus far, and our research roadmap.more » « less
-
Abstract Protein labeling strategies have been explored for decades to study protein structure, function, and regulation. Fluorescent labeling of a protein enables the study of protein‐protein interactions through biophysical methods such as microscale thermophoresis (MST). MST measures the directed motion of a fluorescently labeled protein in response to microscopic temperature gradients, and the protein's thermal mobility can be used to determine binding affinity. However, the stoichiometry and site specificity of fluorescent labeling are hard to control, and heterogeneous labeling can generate inaccuracies in binding measurements. Here, we describe an easy‐to‐apply protocol for high‐stoichiometric, site‐specific labeling of a protein at its N‐terminus with
N ‐hydroxysuccinimide (NHS) esters as a means to measure protein‐protein interaction affinity by MST. This protocol includes guidelines for NHS ester labeling, fluorescent‐labeled protein purification, and MST measurement using a labeled protein. As an example of the entire workflow, we additionally provide a protocol for labeling a ubiquitin E3 enzyme and testing ubiquitin E2‐E3 enzyme binding affinity. These methods are highly adaptable and can be extended for protein interaction studies in various biological and biochemical circumstances. © 2021 Wiley Periodicals LLC.This article was corrected on 18 July 2022. See the end of the full text for details.
Basic Protocol 1 : Labeling a protein of interest at its N‐terminus with NHS esters through stepwise reactionAlternate Protocol : Labeling a protein of interest at its N‐terminus with NHS esters through a one‐pot reactionBasic Protocol 2 : Purifying the N‐terminal fluorescent‐labeled protein and determining its concentration and labeling efficiencyBasic Protocol 3 : Using MST to determine the binding affinity of an N‐terminal fluorescent‐labeled protein to a binding partner.Basic Protocol 4 : NHS ester labeling of ubiquitin E3 ligase WWP2 and measurement of the binding affinity between WWP2 and an E2 conjugating enzyme by the MST binding assay -
Subgoal labeling is an instructional design framework for breaking down problems into pieces that are small enough for novices to grasp, and often difficult for instructors (i.e., experts) to articulate. Subgoal labels have been shown to improve student performance during problem solving in many disciplines, including computing. Improved student performance occurs because subgoal labels improve student transfer and retention of knowledge. With support from NSF (DUE-1712025, 1712231, 1927906, 2110156, 2111578), subgoal labels have been previously identified and integrated into a CS1 course (variables, expressions, conditionals, loops, arrays, classes) and an e-book has been created on the Runestone platform to enable students to complete practice problems using the subgoals. The initial implementation focused on Java, but within the past year, the development of subgoals for CS1 courses in Python have been created. Subsequently, course materials have been created as well. This workshop will introduce participants to the new materials (in Python) and demonstrate how the subgoal labels and worked examples are integrated throughout the course. Materials include worked examples and practice problems that increase in complexity and difficulty within each topic. The materials are designed to be integrated into CS1 courses as homework or classroom examples and activities. Assessment of topics using subgoal labels will also be discussed. Participants will also engage in an activity where they create an example for their own course using subgoal labels.more » « less
-
Abstract Studying facial expressions is a notoriously difficult endeavor. Recent advances in the field of affective computing have yielded impressive progress in automatically detecting facial expressions from pictures and videos. However, much of this work has yet to be widely disseminated in social science domains such as psychology. Current state-of-the-art models require considerable domain expertise that is not traditionally incorporated into social science training programs. Furthermore, there is a notable absence of user-friendly and open-source software that provides a comprehensive set of tools and functions that support facial expression research. In this paper, we introduce Py-Feat, an open-source Python toolbox that provides support for detecting, preprocessing, analyzing, and visualizing facial expression data. Py-Feat makes it easy for domain experts to disseminate and benchmark computer vision models and also for end users to quickly process, analyze, and visualize face expression data. We hope this platform will facilitate increased use of facial expression data in human behavior research.