paper-with-me

홈 › Papers

MiLA: Multi-view Intensive-fidelity Long-term Video Generation World Model for Autonomous Driving

2025-03-20 · Haiguang Wang, Daqi Liu, Hongwei Xie, Haisong Liu, Enhui Ma, Kaicheng Yu, LiMin Wang, Bing Wang

In recent years, data-driven techniques have greatly advanced autonomous driving systems, but the need for rare and diverse training data remains a challenge, requiring significant investment in equipment and labor. World models, which predict and generate future environmental states, offer a promising solution by synthesizing annotated video data for training. However, existing methods struggle to generate long, consistent videos without accumulating errors, especially in dynamic scenes. To address this, we propose MiLA, a novel framework for generating high-fidelity, long-duration videos up to one minute. MiLA utilizes a Coarse-to-Re(fine) approach to both stabilize video generation and correct distortion of dynamic objects. Additionally, we introduce a Temporal Progressive Denoising Scheduler and Joint Denoising and Correcting Flow modules to improve the quality of generated videos. Extensive experiments on the nuScenes dataset show that MiLA achieves state-of-the-art performance in video generation quality. For more information, visit the project website: https://github.com/xiaomi-mlab/mila.github.io.

📄 PDF Abstract BibTeX arXiv:2503.15875

Code (1)

xiaomi-mlab/mila.github.io 공식 구현

Tasks

Autonomous DrivingDenoisingVideo Generation

Similar Papers 제목 키워드 기반

Refining Few-Step Text-to-Multiview Diffusion via Reinforcement Learning

2025-05-26 · Ziyi Zhang, Li Shen, Deheng Ye, Yong Luo 외

Text-to-multiview (T2MV) generation, which produces coherent multiview images from a single text prompt, remains computationally intensive, while accelerated T2MV methods using few-step diffusion models often sacrifice i…

Denoisingreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Easy3E: Feed-Forward 3D Asset Editing via Rectified Voxel Flow

2026-02-25 · Shimin Hu, Yuanyi Wei, Fei Zha, Yudong Guo 외 arxiv

Existing 3D editing methods rely on computationally intensive scene-by-scene iterative optimization and suffer from multi-view inconsistency. We propose an effective and feed-forward 3D editing framework based on the TRE…

Overview of Gaussian process based multi-fidelity techniques with variable relationship between fidelities

2020-06-30 · Loïc Brevault, Mathieu Balesdent, Ali Hebbal

The design process of complex systems such as new configurations of aircraft or launch vehicles is usually decomposed in different phases which are characterized for instance by the depth of the analyses in terms of numb…

Gaussian Processes

Beyond Semantic Similarity: A Two-Phase Non-Parametric Retrieval Workflow for Corporate Credit Underwriting

2026-05-20 · Linus Ng Junjia, Ezekiel Tee Kongquan, Kelvin Heng, Kenneth Zhu Ke 외 arxiv

Corporate credit underwriting requires analysts to extract actionable evidence from long, heterogeneous financial documents spanning hundreds of pages and multiple languages. Standard Retrieval-Augmented Generation (RAG)…

Semantic Similarity

Enhanced Diagnostic Fidelity in Pathology Whole Slide Image Compression via Deep Learning

2025-03-14 · Maximilian Fischer, Peter Neher, Peter Schüffler, Shuhan Xiao 외

Accurate diagnosis of disease often depends on the exhaustive examination of Whole Slide Images (WSI) at microscopic resolution. Efficient handling of these data-intensive images requires lossy compression techniques. Th…

DiagnosticImage CompressionMS-SSIMSSIM+1