About
Caleb is a Taiwanese-Chinese Texan, age 5 (likely tall for his age). He researches useful, energy-efficient AI tech alongside the wizards at Latent AI (the wizards took pity on him). He is particularly fond of consuming Chick-fil-A sandwiches and overusing the egg 🥚 emoji. He lives in Austin with his Squishmallow, Gustavus.
He writes Tung-In-Cheek over at 
    
        
            Substack
        
        
    
Publications
An Automated Approach For Improving The Inference Latency And Energy Efficiency Of Pretrained CNNs By Removing Irrelevant Pixels With Focused Convolutions 
2024 IEEE Asia and South Pacific Design Automation Conference 
C Tung, N Eliopoulos, P Jajal, G Ramshankar, C-Y Yang, N Synovic, X Zhang, V Chaudhary, GK Thiruvathukal, and Y-H Lu
Directed Acyclic Graph-based Neural Networks for Tunable Low-Power Computer Vision 
2022 ACM/IEEE International Symposium on Low Power Electronics and Design 
A Goel, C Tung, N Eliopoulos, X Hu, GK Thiruvathukal, JC Davis, and Y-H Lu
Efficient Computer Vision on Edge Devices with Pipeline-Parallel Hierarchical Neural Networks 
2022 IEEE Asia and South Pacific Design Automation Conference 
A Goel, C Tung, X Hu, GK Thiruvathukal, JC Davis, and Y-H Lu
Irrelevant Pixels are Everywhere: Find and Exclude Them for More Efficient Computer Vision 
2022 IEEE International Conference on Artificial Intelligence Circuits and Systems 
C Tung, A Goel, N Eliopoulos, X Hu, GK Thiruvathukal, V Chaudhary, and Y-H Lu
Why Accuracy Is Not Enough: The Need for Consistency in Object Detection 
2022 IEEE Multimedia 
C Tung, A Goel, F Bordwell, N Eliopoulos, X Hu, GK Thiruvathukal, and Y-H Lu
Observing Human Mobility Worldwide during COVID-19 
2022 IEEE Computer 
S Allcroft, M Metwaly, Z Berg, I Ghodgaonkar, F Bordwell, X Zhao, X Liu, S Chakraborty, V Banna, A Chinnakotla, A Goel, C Tung, G Kao, W Zakharov, D Shoham, GK Thiruvathukal, and Y-H Lu
Low-Power Multi-Camera Object Re-Identification using Hierarchical Neural Networks 
2021 ACM/IEEE International Symposium on Low Power Electronics and Design 
A Goel, C Tung, X Hu, H Wang, JC Davis, GK Thiruvathukal and Y-H Lu
Modular Neural Networks for Low-Power Image Classification on Embedded Devices 
2020 ACM Transactions on Design Automation of Electronic Systems 
A Goel, C Tung, S Aghajanzadeh, S-H Chen, GK Thiruvathukal, Y-H Lu
A Survey of Methods for Low-Power Deep Learning and Computer Vision 
2020 IEEE World Forum on Internet of Things 
A Goel, C Tung, Y-H Lu, GK Thiruvathukal
Analyzing Worldwide Social Distancing through Large-Scale Computer Vision 
2020 ArXiV 
I Ghodgaonkar, S Chakraborty, V Banna, S Allcroft, M Metwaly, F Bordwell, K Kimura, X Zhao, A Goel, C Tung, A Chinnakotla, M Xue, Y-H Lu, MD Ward, W Zakharov, DS Ebert, DM Barbarash, GK Thiruvathukal
Low-Power Object Counting with Hierarchical Neural Networks 
2020 ACM/IEEE International Symposium on Low Power Electronics and Design 
A Goel, C Tung, S Aghajanzadeh, I Ghodgaonkar, S Ghosh, GK Thiruvathukal, Y-H Lu
Camera Placement Meeting Restrictions of Computer Vision 
2020 IEEE International Conference on Image Processing 
S Aghajanzadeh, R Naidu, S-H Chen, C Tung, A Goel, Y-H Lu, GK Thiruvathukal
Large-Scale Object Detection of Images from Network Cameras in Variable Ambient Lighting Conditions 
2019 IEEE International Conference on Multimedia Information Processing & Retrieval 
C Tung, MR Kelleher, RJ Schlueter, B Xu, Y-H Lu, GK Thiruvathukal, Y-K Chen, Y Lu
Internships
More details on 
    
        
            LinkedIn
        
        
    
2022 @ 
    
        
            Arm
        
        
    
 
efficient neural network architectures
2021 @ 
    
        
            Latent AI
        
        
    
 
dynamic, throttlable neural network prototyping
2019 @ 
    
        
            Southwest Research Institute
        
        
    
 
drone computer vision for methane detection
2018 @ 
    
        
            USAA
        
        
    
 
web browser tools for developers
2017 @ 
    
        
            Purdue University HELPS Lab
        
        
    
 
supercomputer-powered object recognition
2016 @ 
    
        
            Software Engineering Professionals
        
        
    
 
Windows application frontend
 
         
                        
                        
                     
                        
                        
                    