✨ShotStream: Streaming Multi-Shot Video Generation for Interactive Storytelling

📝 Summary:
ShotStream enables real-time interactive multi-shot video generation via a novel causal architecture. It uses dual-cache memory for visual consistency and two-stage distillation to reduce latency and error. This achieves high-quality, coherent videos at 16 FPS, paving the way for dynamic storytel...

🔹 Publication Date: Published on Mar 26

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.25746
• PDF: https://arxiv.org/pdf/2603.25746
• Project Page: https://luo0207.github.io/ShotStream/
• Github: https://github.com/KlingAIResearch/ShotStream

🔹 Models citing this paper:
• https://huggingface.co/KlingTeam/ShotStream

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#VideoGeneration #GenerativeAI #RealTimeAI #DeepLearning #AIStorytelling

✨PackForcing: Short Video Training Suffices for Long Video Sampling and Long Context Inference

📝 Summary:
PackForcing enables efficient, long-video generation via hierarchical KV-cache management and spatiotemporal compression, overcoming memory and consistency issues. It generates 2-minute coherent videos on a single GPU, demonstrating that short-video training suffices for high-quality long-video s...

🔹 Publication Date: Published on Mar 26

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.25730
• PDF: https://arxiv.org/pdf/2603.25730
• Github: https://github.com/ShandaAI/PackForcing

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#VideoGeneration #GenerativeAI #DeepLearning #ModelEfficiency #LongContext

✨TokenDial: Continuous Attribute Control in Text-to-Video via Spatiotemporal Token Offsets

📝 Summary:
TokenDial enables precise attribute control in text-to-video models by using additive offsets in spatiotemporal token space for coherent edits without retraining. AI-generated summary We present Token...

🔹 Publication Date: Published on Mar 29

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.27520
• PDF: https://arxiv.org/pdf/2603.27520
• Project Page: https://tokendial.github.io/
• Github: https://github.com/ariannaliu/TokenDial

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#TextToVideo #GenerativeAI #AIControl #VideoGeneration #DeepLearning

✨DynaVid: Learning to Generate Highly Dynamic Videos using Synthetic Motion Data

📝 Summary:
DynaVid improves dynamic video synthesis by training with synthetic optical flow, which provides diverse motion patterns without artificial appearances. A two-stage framework learns dynamic motion while preserving visual realism, enhancing motion control.

🔹 Publication Date: Published on Apr 2

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2604.01666
• PDF: https://arxiv.org/pdf/2604.01666
• Project Page: https://jinwonjoon.github.io/DynaVid/

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#VideoGeneration #AIVideo #DeepLearning #ComputerVision #SyntheticData

✨Salt: Self-Consistent Distribution Matching with Cache-Aware Training for Fast Video Generation

📝 Summary:
The paper introduces Salt, a method for fast video generation. It proposes Self-Consistent Distribution Matching Distillation SC-DMD to improve low-NFE quality by regularizing denoising updates. Cache-Distribution-Aware training further optimizes real-time autoregressive generation using KV cache.

🔹 Publication Date: Published on Apr 3

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2604.03118
• PDF: https://arxiv.org/pdf/2604.03118
• Github: https://github.com/XingtongGe/Salt

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#VideoGeneration #GenerativeAI #DeepLearning #AIResearch #RealTimeAI

✨Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memory

📝 Summary:
Matrix-Game 3.0 is a memory-augmented diffusion model achieving real-time 720p interactive video generation with long-term temporal consistency. It uses an advanced data engine, a self-correction training framework with memory, and efficient inference strategies. This enables practical, industria...

🔹 Publication Date: Published on Apr 10

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2604.08995
• PDF: https://arxiv.org/pdf/2604.08995
• Project Page: https://matrix-game-v3.github.io/

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#DiffusionModels #VideoGeneration #RealTimeAI #GenerativeAI #MachineLearning

✨CT-1: Vision-Language-Camera Models Transfer Spatial Reasoning Knowledge to Camera-Controllable Video Generation

📝 Summary:
CT-1 is a Vision-Language-Camera model that improves camera-controllable video generation. It uses a Diffusion Transformer and Wavelet Regularization Loss to accurately estimate camera trajectories, enabling precise video synthesis. This achieves 25.7% better accuracy than prior methods.

🔹 Publication Date: Published on Apr 10

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2604.09201
• PDF: https://arxiv.org/pdf/2604.09201
• Project Page: https://gulucaptain.github.io/Camera-Transformer-1/
• Github: https://github.com/gulucaptain/Camera-Transformer-1

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#AI #VideoGeneration #ComputerVision #DiffusionModels #VisionLanguageModels

✨Uni-ViGU: Towards Unified Video Generation and Understanding via A Diffusion-Based Video Generator

📝 Summary:
Uni-ViGU introduces a unified framework for video generation and understanding, uniquely building upon a video generator as its foundation. It uses unified flow matching and a bidirectional training mechanism to achieve competitive performance in both generation and understanding tasks.

🔹 Publication Date: Published on Apr 9

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2604.08121
• PDF: https://arxiv.org/pdf/2604.08121
• Project Page: https://fr0zencrane.github.io/uni-vigu-page/
• Github: https://fr0zencrane.github.io/uni-vigu-page/

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#VideoGeneration #VideoUnderstanding #DiffusionModels #AIResearch #DeepLearning

✨CityRAG: Stepping Into a City via Spatially-Grounded Video Generation

📝 Summary:
CityRAG generates long-term, physically grounded video sequences that maintain environmental consistency and support complex navigation through real-world geography using geo-registered data as contex...

🔹 Publication Date: Published on Apr 21

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2604.19741
• PDF: https://arxiv.org/pdf/2604.19741
• Project Page: https://cityrag.github.io/

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#VideoGeneration #GenerativeAI #SpatialAI #ComputerVision #UrbanSimulation

🔥 Warp-as-History: Generalizable Camera-Controlled Video Generation from One Training Video

💡 The paper proposes a novel approach called Warp-as-History for camera-controlled video generation. Existing methods for this task typically require large-scale camera-annotated videos for post-training or rely on test-time optimization, which can be time-consuming and costly. The proposed method addresses this problem by transforming camera-induced warps into pseudo-history representations, which enables a frozen video generation model to follow camera trajectories without any training or test-time optimization.

The Warp-as-History method works by constructing camera-warped pseudo-history from past observations and feeding it through the model's visual-history pathway. The positional encoding is aligned with the target frames being denoised, and warped-history tokens without valid source observations are removed. This simple interface reveals a non-trivial zero-shot capability of the model to follow camera trajectories.

The results show that the proposed method can achieve good camera adherence, visual quality, and motion dynamics without requiring large-scale camera-annotated videos or test-time optimization. Furthermore, lightweight offline finetuning on only one camera-annotated video can further improve the model's capability and generalize to unseen videos. Extensive experiments on diverse datasets confirm the effectiveness of the Warp-as-History method, making it a promising approach for camera-controlled video generation.

Overall, the paper's contributions include a novel method for camera-controlled video generation that requires minimal training data and no test-time optimization, and demonstrates the potential for zero-shot capability in video generation models. The proposed approach has the potential to simplify the process of camera-controlled video generation and make it more accessible to a wider range of applications.

📅 Published on May 14

🔗 Links:
• GitHub: https://github.com/huggingface
• arXiv: https://arxiv.org/abs/2605.15182
• PDF: https://arxiv.org/pdf/2605.15182
• Project Page: https://yyfz.github.io/warp-as-history/

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#VideoGeneration #CameraControlledSynthesis #WarpAsHistory #PseudoHistoryRepresentations #CameraInducedWarps