Chen-yi Lu
About Me
I am a 3rd-year Ph.D. student at Purdue University, advised by Prof. Somali Chaterji.
My research focuses on adversarially robust and data-efficient learning algorithms for computer vision and multimodal tasks. I have worked on problems spanning image semantic segmentation, video understanding, text-to-image diffusion models, and video coherence metrics, with broader interests in generative modeling and robust multimodal learning.
In the summers of 2023 and 2024, I interned at Adobe Research, collaborating with Mehrab Tanjim. At Adobe, I developed methods for retrieval-augmented diffusion to improve inference efficiency, and designed a learned metric for multi-shot video coherence.
Before joining Purdue, I was a Research Assistant at National Taiwan University (NTU) with Prof. Ta-Te Lin. My work focused on precision agriculture, where I built embedded systems and ML algorithms for interpretable decision support, robust edge deployment, and real-world agricultural monitoring.
News
Selected Publications (Full list on Google Scholar)
-
ECCV
What CLIP Knows but Cannot Say: Recovering Negation from Frozen Intermediate FeaturesEuropean Conference on Computer Vision (ECCV), 2026. -
ICCV
SKALD: Learning-Based Shot Assembly for Coherent Multi-Shot Video CreationInternational Conference on Computer Vision, 2025. -
CVPR
Improving Semi-supervised Semantic Segmentation with Sliced-Wasserstein Feature Alignment and UniformityThe IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025. -
ECCV
ReCon: Training-Free Acceleration for Text-to-Image Synthesis with Retrieval of Concept Prompt TrajectoriesEuropean Conference on Computer Vision (ECCV), 2024. -
Biosystems Engineering
Online semi-supervised learning applied to an automated insect pest monitoring systemBiosystems Engineering, 2021, 208, 28-44 -
Pest Management Science
Towards intelligent and integrated pest management through an AIoTābased monitoring systemPest Management Science, 2022, 78 (10), 4288-4302 -
Computers & Electronics in Ag.
Edge-based wireless imaging system for continuous monitoring of insect pests in a remote outdoor mango orchardComputers and Electronics in Agriculture, 2023, 211, 108019 -
IFAC-PapersOnLine
Generative adversarial network based image augmentation for insect pest classification enhancementIFAC-PapersOnLine, 2019, 52 (30), 1-5
Powered by Jekyll and Minimal Light theme.