Meta tags:
keywords=computer vision,machine learning,researcher;
description=Student researcher;
author=Me;

Headings (most frequently used words):

for, video, generation, visual, via, and, learning, 2022, aware, active, semantic, segmentation, lee, hsin, ying, 李信穎, selected, publications, motimotion, motion, controlled, with, reasoning, icml, 2026, consistent, subject, contrastive, instantiated, concepts, tmlr, 2025, exploiting, diffusion, prior, generalizable, dense, prediction, cvpr, 2024, crossdtr, cross, view, depth, guided, transformers, 3d, object, detection, icra, 2023, fine, grained, understanding, question, answering, decoupling, spatial, temporal, modeling, bmvc, spotlight, d2ada, dynamic, density, domain, adaptation, eccv, redal, region, based, diversity, point, cloud, iccv, 2021,

Text of the page (most frequently used words):
arxiv (14), lee (12), hsin (12), ying (12), video (6), winston (5), hsu (5), for (5), tsung (4), han (4), and (4), chen (4), page (4), visual (4), ming (4), hsuan (4), yang (4), huang (3), hung (3), #learning (3), code (3), #generation (3), 2026 (2), ting (2), aware (2), active (2), semantic (2), segmentation (2), 2022 (2), language (2), via (2), tseng (2), generating (2), consistent (2), motion (2), reasoning (2), with (2), publications (2), phd (2), prof (2), university (2), powered, papermod, hugo, kai, ping, chia, iccv, 2021, yueh, cheng, liu, redal, region, based, diversity, point, cloud, syuan, liou, shao, yuan, kuan, chih, eccv, tung, d2ada, dynamic, density, domain, adaptation, bing, tsai, bmvc, spotlight, jia, fong, yeh, double, stream, encoder, fine, grained, understanding, question, answering, decoupling, spatial, temporal, modeling, ching, rong, wen, chin, icra, 2023, crossdtr, cross, view, depth, guided, transformers, object, detection, cvpr, 2024, universal, transferring, framework, exploiting, diffusion, prior, generalizable, dense, prediction, tmlr, 2025, kelvin, chan, subjects, across, multiple, creations, subject, contrastive, instantiated, concepts, icml, zhixin, shu, jing, shi, yiqun, mei, hanwen, jiang, plausible, over, intent, causality, motimotion, controlled, selected, before, journey, also, worked, national, taiwan, second, year, student, the, california, merced, advised, current, research, topics, are, image, interested, developing, interactable, intelligence, 李信穎,

Text of the page (random words):
lee hsin ying lee hsin ying publications lee hsin ying 李信穎 i am a second year phd student at the university of california merced advised by prof ming hsuan yang my current research topics are in image and video generation i m interested in developing interactable visual intelligence before my phd journey i also worked on visual language learning with prof winston hsu at national taiwan university selected publications motimotion motion controlled video generation with visual reasoning generating plausible motion by reasoning over intent and causality lee hsin ying hanwen jiang yiqun mei jing shi ming hsuan yang zhixin shu icml 2026 page arxiv arxiv consistent subject generation via contrastive instantiated concepts generating consistent subjects across multiple creations lee hsin ying kelvin c k chan ming hsuan yang tmlr 2025 page arxiv arxiv code exploiting diffusion prior for generalizable dense prediction a universal transferring framework lee hsin ying hung yu tseng hsin ying lee ming hsuan yang cvpr 2024 page arxiv arxiv code video crossdtr cross view and depth guided transformers for 3d object detection ching yu tseng yi rong chen lee hsin ying tsung han wu wen chin chen winston hsu icra 2023 arxiv arxiv learning fine grained visual understanding for video question answering via decoupling spatial temporal modeling a double stream video language encoder lee hsin ying hung ting su bing chen tsai tsung han wu jia fong yeh winston hsu bmvc 2022 spotlight page arxiv arxiv code video d2ada dynamic density aware active domain adaptation for semantic segmentation tsung han wu yi syuan liou shao ji yuan lee hsin ying tung i chen kuan chih huang winston hsu eccv 2022 arxiv arxiv redal region based and diversity aware active learning for point cloud semantic segmentation tsung han wu yueh cheng liu yu kai huang lee hsin ying hung ting su ping chia huang winston h hsu iccv 2021 arxiv arxiv 2026 lee hsin ying powered by hugo papermod