Meta tags:
author=Muyao Niu;

Headings (most frequently used words):

selected, papers, full, publication, list, chronological, awards, scholarships,

Text of the page (most frequently used words):
the (45), muyao (29), niu (29), and (28), paper (26), video (26), code (23), zheng (23), yinqiang (22), for (21), human (18), model (14), yifan (14), zhan (14), nir (13), zhihang (13), zhong (13), from (12), wang (12), page (11), image (10), project (10), zhang (10), diffusion (10), with (9), zhuoxiao (9), via (9), aware (9), motion (9), qingtian (9), zhu (9), 2026 (9), 2024 (8), based (8), 2023 (7), framework (7), chen (7), videos (6), region (6), enhancement (6), imaging (6), images (6), eccv (6), models (6), zhao (6), mofa (6), avatars (6), xiao (6), sun (6), tree (6), gaussian (6), wei (6), avatar (6), rgb (6), network (5), which (5), perception (5), adversarial (5), that (5), acm (5), vae (5), introduce (5), generation (5), 2025 (5), mingze (5), university (4), multi (4), liu (4), illumination (4), spectrum (4), dark (4), vis (4), vision (4), visual (4), cvpr (4), data (4), nerf (4), neural (4), enabling (4), scene (4), generative (4), yong (4), xiaodong (4), cun (4), ying (4), shan (4), animation (4), frozen (4), animate (4), control (4), long (4), novel (4), arxiv (4), 3dgs (4), under (4), surprise (4), darkfusion (4), robust (4), background (4), research (3), year (3), tokyo (3), transformation (3), propose (3), assisted (3), sketch (3), proposed (3), method (3), information (3), design (3), designed (3), range (3), performance (3), low (3), light (3), using (3), iccv (3), physics (3), innovative (3), approach (3), field (3), sampling (3), xiang (3), latent (3), uniform (3), yang (3), single (3), gao (3), jiancheng (3), nerv (3), weihang (3), ran (3), guo (3), spacedg (3), special (2), scholarship (2), 000 (2), jpy (2), scale (2), learning (2), ning (2), zhihui (2), zhiyong (2), bin (2), haojie (2), colorization (2), uses (2), within (2), visibility (2), constrained (2), wide (2), band (2), seeing (2), optimal (2), considering (2), constraints (2), significantly (2), improves (2), translation (2), fully (2), differentiable (2), was (2), includes (2), process (2), improve (2), pseudo (2), trained (2), attack (2), camera (2), systems (2), developed (2), powered (2), detection (2), dynamic (2), kalman (2), radiance (2), reconstruction (2), compatible (2), our (2), sijie (2), xiaoyu (2), wenbo (2), controllable (2), adaptions (2), adapt (2), motions (2), different (2), domains (2), can (2), effectively (2), various (2), types (2), signals (2), including (2), trajectories (2), keypoint (2), sequences (2), their (2), combinations (2), xintao (2), high (2), fidelity (2), stereoscopic (2), immersive (2), event (2), pedestrian (2), end (2), clothing (2), yin (2), explicit (2), exoskeleton (2), complicated (2), most (2), modeling (2), structured (2), representation (2), efficient (2), non (2), encoding (2), feature (2), zunian (2), wan (2), compression (2), webgpu (2), splatting (2), platform (2), yuning (2), gong (2), yifei (2), yuanyuan (2), zhou (2), xue (2), mad (2), animatable (2), deblurring (2), deriving (2), sharp (2), intrinsic (2), blurry (2), blur (2), formation (2), driving (2), spatial (2), intelligence (2), degradation (2), forcing (2), toward (2), fusion (2), leverages (2), complementary (2), achieve (2), extremely (2), conditions (2), chang (2), wen (2), anicrafter (2), customizing (2), realistic (2), centric (2), conditioning (2), leverage (2), guidance (2), insert (2), anyone (2), into (2), any (2), following (2), given (2), sequence (2), mingdeng (2), cao (2), yanhong (2), zeng (2), 牛慕尧 (2), template, comes, personal, website, jon, barron, boost, nais, 900, wing, cfs, assistant, 160, 2027, awards, scholarships, coloring, anime, line, art, tre, net, enhance, geometric, regions, pattern, recognition, zhi, dou, zhaoyan, ming, rasc, map, better, utilize, regional, enhancing, wise, features, tip, kun, unpaired, hour, addressed, issue, collecting, utilizing, physiscs, inspired, algorithms, are, simulate, paired, simulating, day, night, situations, then, generated, near, infrared, detector, nighttime, surveillance, introduced, passively, manipulates, intensity, distribution, black, box, algorithm, target, deep, huy, nguyen, isao, echizen, kfd, rethinking, filter, combine, filtering, accurate, deformation, estimation, observations, predictions, shohei, nobuhara, nishino, fields, rolling, shutter, consider, distortions, two, technologies, trajectory, smoothness, regularization, strategy, tong, existing, provides, truly, spatio, temporally, compressed, space, opposed, frame, neurips, shaoshu, stereocrafter, monocular, present, converting, addressing, growing, demand, content, experience, leveraging, foundation, priors, boosts, ensure, required, display, devices, technical, report, zhe, kong, xiangjun, attacks, detectors, physical, driven, framing, textures, texture, optimization, problem, aaai, guixu, lin, zhengwei, shengfeng, tomie, towards, highlight, critical, yet, often, overlooked, factor, tasks, namely, hand, held, objects, loose, fitting, body, qiao, adaptive, unlike, conventional, approaches, organizes, representations, binary, search, bst, along, temporal, axis, all, one, transferring, machine, asymmetric, adaptor, bridges, learned, task, compact, cheng, ching, tseng, visionary, world, carrier, built, build, open, onnx, runtime, real, time, rendering, diverse, variants, mlp, 4dgs, xueying, yuanjun, liao, jiaming, jiaqi, minming, xiaoqing, hou, huaxi, huang, shixiang, tang, dingwen, junchi, yan, yanchi, composing, worlds, through, disentangled, scenario, composia, powerful, simulator, synthesizing, rare, scenes, over, structure, identity, ego, action, zhengqing, qingjie, zhuo, xiaoyang, weiqiang, ren, qian, benchmarking, first, large, dataset, bench, verified, benchmark, evaluating, mllms, degradations, xiaolong, ziyang, jiarui, qiyue, hongjie, what, remember, when, skip, training, free, keep, informative, frames, discarded, sliding, window, cache, memory, bank, adaptively, cut, denoising, steps, easy, chunks, preserving, consistency, narrative, coherence, autoregressive, shuwei, shi, zhen, chuanhao, kaipeng, full, publication, list, chronological, selected, papers, wechat, myniuuu, google, scholar, github, email, 2nd, student, department, mechano, informatics, utokyo, supervisor, prof, received, degree, dalian, technology, 2022, interests, include, aigc, computational, photography,

Text of the page (random words):
muyao niu 牛慕尧 muyao niu 牛慕尧 hi i m muyao niu i am a 2nd year ph d student in the department of mechano informatics the university of tokyo utokyo my supervisor is prof yinqiang zheng i received my b e degree from dalian university of technology in 2022 my research interests include aigc computational photography and 3d vision email github google scholar wechat myniuuu selected 5 papers mofa video controllable image animation via generative motion field adaptions in frozen image to video diffusion model muyao niu xiaodong cun xintao wang yong zhang ying shan yinqiang zheng eccv 2024 project page paper code we introduce mofa video to adapt motions from different domains to the frozen video diffusion model mofa video can effectively animate a single image using various types of control signals including trajectories keypoint sequences and their combinations anicrafter customizing realistic human centric animation via avatar background conditioning in video diffusion models muyao niu mingdeng cao yifan zhan qingtian zhu weihang ran yanhong zeng xiao sun zhihang zhong yinqiang zheng acm mm 2026 project page paper code we leverage 3dgs avatar background video as guidance for the video diffusion model to insert and animate anyone into any scene following given motion sequence mad avatar motion aware animatable gaussian avatars deblurring muyao niu yifan zhan qingtian zhu zhuoxiao li wei wang zhihang zhong xiao sun yinqiang zheng cvpr 2026 paper code we introduce an innovative method for deriving sharp intrinsic 3d human gaussian avatars from blurry videos via 3d human motion model and the physics based blur formation model darkfusion toward robust and 3d aware rgb nir imaging in the dark muyao niu mingze ma yifan zhan qingtian zhu zhihang zhong wei guo chang wen chen yinqiang zheng acm mm 2026 paper code darkfusion is a novel rgb nir fusion framework that leverages the complementary information from rgb and nir images to achieve robust and 3d aware imaging under extremely low light conditions visibility constrained wide band illumination spectrum design for seeing in the dark muyao niu zhuoxiao li zhihang zhong yinqiang zheng cvpr 2023 paper code we designed an optimal illumination spectrum in the vis nir range by considering human vision constraints which significantly improves translation performance a fully differentiable model was proposed which includes the imaging process human visual perception and the enhancement network full publication list chronological anicrafter customizing realistic human centric animation via avatar background conditioning in video diffusion models muyao niu mingdeng cao yifan zhan qingtian zhu weihang ran yanhong zeng xiao sun zhihang zhong yinqiang zheng acm mm 2026 project page paper code we leverage 3dgs avatar background video as guidance for the video diffusion model to insert and animate anyone into any scene following given motion sequence darkfusion toward robust and 3d aware rgb nir imaging in the dark muyao niu mingze ma yifan zhan qingtian zhu zhihang zhong wei guo chang wen chen yinqiang zheng acm mm 2026 paper code darkfusion is a novel rgb nir fusion framework that leverages the complementary information from rgb and nir images to achieve robust and 3d aware imaging under extremely low light conditions surprise forcing what to remember when to skip in long video generation shuwei shi zhen li muyao niu chuanhao li bo zheng kaipeng zhang yinqiang zheng eccv 2026 project page paper code surprise forcing is a training free framework that uses surprise to keep the most informative frames discarded by the sliding window kv cache via a surprise aware memory bank and to adaptively cut denoising steps for easy chunks preserving long range scene consistency and narrative coherence in autoregressive diffusion long video generation spacedg benchmarking spatial intelligence under visual degradation xiaolong zhou yifei liu ziyang gong jiarui li qiyue zhao muyao niu yuanyuan gao le ma xue yang hongjie zhang zhihang zhong arxiv 2026 project page paper code we introduce spacedg the first large scale dataset for degradation aware spatial intelligence and spacedg bench a human verified benchmark for evaluating mllms under visual degradations composing driving worlds through disentangled control for adversarial scenario generation yifan zhan zhengqing chen qingjie wang zhuo he muyao niu xiaoyang guo wei yin weiqiang ren qian zhang yinqiang zheng eccv 2026 project page paper code composia is a powerful simulator for synthesizing rare driving scenes enabling explicit control over structure identity and ego action mad avatar motion aware animatable gaussian avatars deblurring muyao niu yifan zhan qingtian zhu zhuoxiao li wei wang zhihang zhong xiao sun yinqiang zheng cvpr 2026 paper code we introduce an innovative method for deriving sharp intrinsic 3d human gaussian avatars from blurry videos via 3d human motion model and the physics based blur formation model visionary the world model carrier built on webgpu powered gaussian splatting platform yuning gong yifei liu yifan zhan muyao niu xueying li yuanjun liao jiaming chen yuanyuan gao jiaqi chen minming chen li zhou yuning zhang wei wang xiaoqing hou huaxi huang shixiang tang le ma dingwen zhang xue yang junchi yan yanchi zhang yinqiang zheng xiao sun zhihang zhong arxiv 2025 project page paper code we build an open platform via webgpu and onnx runtime enabling real time rendering of diverse gaussian splatting variants 3dgs mlp based 3dgs 4dgs neural avatars all in one transferring image compression from human perception to multi machine perception jiancheng zhao xiang ji zhuoxiao li zunian wan weihang ran mingze ma muyao niu yifan zhan cheng ching tseng yinqiang zheng arxiv 2025 paper code an asymmetric adaptor framework bridges learned image compression and multi task vision with a single compact model tree nerv a tree structured neural representation for efficient non uniform video encoding jiancheng zhao yifan zhan qingtian zhu mingze ma muyao niu zunian wan xiang ji yinqiang zheng iccv 2025 paper code we propose tree nerv a novel tree structured feature representation for efficient and adaptive video encoding unlike conventional approaches tree nerv organizes feature representations within a binary search tree bst enabling non uniform sampling along the temporal axis tomie towards explicit exoskeleton for the reconstruction of complicated 3d human avatars yifan zhan qingtian zhu muyao niu mingze ma jiancheng zhao zhihang zhong xiao sun yu qiao yinqiang zheng iccv 2025 paper code we highlight a critical yet often overlooked factor in most 3d human tasks namely modeling complicated 3d human with with hand held objects or loose fitting clothing by modeling the exoskeleton of the human body adversarial attacks on event based pedestrian detectors a physical approach guixu lin muyao niu qingtian zhu zhengwei yin zhuoxiao li shengfeng he yinqiang zheng aaai 2025 paper page code we developed an end to end adversarial framework for event driven pedestrian detection framing the design of adversarial clothing textures as a 2d texture optimization problem stereocrafter diffusion based generation of long and high fidelity stereoscopic 3d from monocular videos sijie zhao wenbo hu xiaodong cun yong zhang xiaoyu li zhe kong xiangjun gao muyao niu ying shan arxiv technical report 2024 project page paper we present a novel framework for converting 2d videos to immersive stereoscopic 3d addressing the growing demand for 3d content in immersive experience leveraging foundation models as priors our approach boosts the performance to ensure the high fidelity generation required by the display devices mofa video controllable image animation via generative motion field adaptions in frozen image to video diffusion model muyao niu xiaodong cun xintao wang yong zhang ying shan yinqiang zheng eccv 2024 project page paper code we introduce mofa video to adapt motions from different domains to the frozen video diffusion model mofa video can effectively animate a single image using various types of control signals including trajectories keypoint sequences and their combinations cv vae a compatible video vae for latent generative video models sijie zhao yong zhang xiaodong cun shaoshu yang muyao niu xiaoyu li wenbo hu ying shan neurips 2024 project page paper code we propose cv vae that is compatible with existing image and video models trained with sd image vae our video vae provides a truly spatio temporally compressed latent space for latent generative video models as opposed to uniform frame sampling rs nerf neural radiance fields from rolling shutter images muyao niu tong chen yifan zhan zhuoxiao li xiang ji yinqiang zheng eccv 2024 paper code we improve nerf to consider the rs distortions with two technologies camera trajectory smoothness regularization and multi sampling strategy kfd nerf rethinking dynamic nerf with kalman filter yifan zhan zhuoxiao li muyao niu zhihang zhong shohei nobuhara ko nishino yinqiang zheng eccv 2024 paper code we combine dynamic neural radiance field with a motion reconstruction framework based on kalman filtering enabling accurate deformation estimation from scene observations and predictions physics based adversarial attack on near infrared human detector for nighttime surveillance camera systems muyao niu zhuoxiao li yifan zhan huy h nguyen isao echizen yinqiang zheng acm mm 2023 paper code we introduced an innovative approach that passively manipulates the intensity distribution of nir images and developed a 3d aware black box attack algorithm to target deep learning based nir powered human detection systems nir assisted video enhancement via unpaired 24 hour data muyao niu zhihang zhong yinqiang zheng iccv 2023 paper code we addressed the issue of collecting data for utilizing nir images to improve low light vis videos physiscs inspired algorithms are designed to simulate pseudo paired data of nir and vis images simulating day to night situations we then trained an enhancement network using the generated pseudo data visibility constrained wide band illumination spectrum design for seeing in the dark muyao niu zhuoxiao li zhihang zhong yinqiang zheng cvpr 2023 paper code we designed an optimal illumination spectrum in the vis nir range by considering human vision constraints which significantly improves translation performance a fully differentiable model was proposed which includes the imaging process human visual perception and the enhancement network region assisted sketch colorization ning wang muyao niu zhihui wang kun hu bin liu zhiyong wang haojie li tip 2023 paper we proposed the region assisted sketch colorization rasc method which uses a region map to better utilize regional information within the sketch enhancing the perception of region wise features coloring anime line art videos with transformation region enhancement network ning wang muyao niu zhi dou zhihui wang zhiyong wang zhaoyan ming bin liu haojie li pattern recognition 2023 paper we propose a multi scale transformation region enhancement network tre net to enhance the learning on geometric transformation regions awards scholarships boost nais special research scholarship 3 900 000 jpy year the university of tokyo 2024 10 2027 09 wing cfs special research assistant scholarship 2 160 000 jpy year the university of tokyo 2023 04 2024 09 the template comes from the personal website of jon barron