18 papers
World-T3: Test-Time Training as Persistent Memory for Real-Time Video World Models
Junyi Wu, Zekun Li, Chenhongyi Yang +9
Recent advances in video world models have enabled plausible interactive environments, yet long-horizon generation places a fundamental demand on visual memory: the previously obse…
Kepler4D: Controllable Future Video Generation via 4D Scene State Evolution
Feiran Wang, Bin Duan, Junyi Wu +2
Video world models aim to preserve scene structure and predict how dynamic objects evolve beyond visual observations. We present Kepler4D, a framework for future video generation t…
Post-training Quantization for Hybrid Iterative Generative Models
Jing Gao, Junyi Wu, Wei Wang +2
Iterative Generative Models (IGMs) span autoregressive and diffusion paradigms, and hybrid variants that couple them can achieve remarkable image-generation fidelity. However, thei…
CogniMap3D: Cognitive 3D Mapping and Rapid Retrieval
Feiran Wang, Junyi Wu, Dawen Cai +2
We present CogniMap3D, a bioinspired framework for dynamic 3D scene understanding and reconstruction that emulates human cognitive processes. Our approach maintains a persistent me…
Consistent Instance Field for Dynamic Scene Understanding
Junyi Wu, Van Nguyen Nguyen, Benjamin Planche +11
We introduce Consistent Instance Field, a continuous and probabilistic spatio-temporal representation for dynamic scene understanding. Unlike prior methods that rely on discrete tr…
TraceFlow: Dynamic 3D Reconstruction of Specular Scenes Driven by Ray Tracing
Jiachen Tao, Junyi Wu, Haoxuan Wang +3
We present TraceFlow, a novel framework for high-fidelity rendering of dynamic specular scenes by addressing two key challenges: precise reflection direction estimation and physica…
From Particles to Fields: Reframing Photon Mapping with Continuous Gaussian Photon Fields
Jiachen Tao, Benjamin Planche, Van Nguyen Nguyen +13
Accurately modeling light transport is essential for realistic image synthesis. Photon mapping provides physically grounded estimates of complex global illumination effects such as…
Motion Marionette: Rethinking Rigid Motion Transfer via Prior Guidance
Haoxuan Wang, Jiachen Tao, Junyi Wu +3
We present Motion Marionette, a zero-shot framework for rigid motion transfer from monocular source videos to single-view target images. Previous works typically employ geometric,…
Orientation-anchored Hyper-Gaussian for 4D Reconstruction from Casual Videos
Junyi Wu, Jiachen Tao, Haoxuan Wang +3
We present Orientation-anchored Gaussian Splatting (OriGS), a novel framework for high-quality 4D reconstruction from casually captured monocular videos. While recent advances exte…
Efficient Multimodal Dataset Distillation via Generative Models
Zhenghao Zhao, Haoxuan Wang, Junyi Wu +3
Dataset distillation aims to synthesize a small dataset from a large dataset, enabling the model trained on it to perform well on the original dataset. With the blooming of large l…
CaO: Rectifying Inconsistencies in Diffusion-Based Dataset Distillation
Haoxuan Wang, Zhenghao Zhao, Junyi Wu +3
The recent introduction of diffusion models in dataset distillation has shown promising potential in creating compact surrogate datasets for large, high-resolution target datasets,…
X-Field: A Physically Grounded Representation for 3D X-ray Reconstruction
Feiran Wang, Jiachen Tao, Junyi Wu +5
X-ray imaging is indispensable in medical diagnostics, yet its use is tightly regulated due to potential health risks. To mitigate radiation exposure, recent research focuses on ge…
Dataset Quantization with Active Learning based Adaptive Sampling
Zhenghao Zhao, Yuzhang Shang, Junyi Wu +1
Deep learning has made remarkable progress recently, largely due to the availability of large, well-labeled datasets. However, the training on such datasets elevates costs and comp…
Visual Grounding with Attention-Driven Constraint Balancing
Weitai Kang, Luowei Zhou, Junyi Wu +2
Unlike Object Detection, Visual Grounding task necessitates the detection of an object described by complex free-form language. To simultaneously model such complex semantic and vi…
PTQ4DiT: Post-training Quantization for Diffusion Transformers
Junyi Wu, Haoxuan Wang, Yuzhang Shang +2
The recent introduction of Diffusion Transformers (DiTs) has demonstrated exceptional capabilities in image generation by using a different backbone architecture, departing from tr…
On the Faithfulness of Vision Transformer Explanations
Junyi Wu, Weitai Kang, Hao Tang +2
To interpret Vision Transformers, post-hoc explanations assign salience scores to input pixels, providing human-understandable heatmaps. However, whether these interpretations refl…
Token Transformation Matters: Towards Faithful Post-hoc Explanation for Vision Transformer
Junyi Wu, Bin Duan, Weitai Kang +2
While Transformers have rapidly gained popularity in various computer vision applications, post-hoc explanations of their internal mechanisms remain largely unexplored. Vision Tran…
QuEST: Low-bit Diffusion Model Quantization via Efficient Selective Finetuning
Haoxuan Wang, Yuzhang Shang, Zhihang Yuan +3
The practical deployment of diffusion models is still hindered by the high memory and computational overhead. Although quantization paves a way for model compression and accelerati…