ScalablePromptus: Scalable and High-Fidelity Prompt-Based Video Streaming

• URL: http://arxiv.org/abs/2607.26106v1
• Date: Tue, 28 Jul 2026 08:59:39 GMT
• Title: ScalablePromptus: Scalable and High-Fidelity Prompt-Based Video Streaming
• Abstract summary: Prompt-based video streaming transmits compact semantic prompts instead of pixel-level content for generative reconstruction.<n>We propose ScalablePromptus, which enhances Promptus with semantic and color-aware prompt inversion, spherical linear for intermediate frames, and--most critically--a dropout training strategy that produces rank-ordered prompt representations.
• Score: 51.48008871921326
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Prompt-based video streaming transmits compact semantic prompts instead of pixel-level content for generative reconstruction, enabling ultra-low-bitrate communication. However, the state-of-the-art Promptus framework is vulnerable to network fluctuation, where partially received prompts lead to catastrophic quality collapse. We propose ScalablePromptus, which enhances Promptus with semantic and color-aware prompt inversion, spherical linear interpolation for intermediate frames, and--most critically--a dropout training strategy that produces rank-ordered prompt representations. This allows the receiver to reconstruct meaningful video from arbitrarily truncated prompts without any adaptation. Under stable networks, ScalablePromptus achieves modest quality gains. Under lossy conditions, it reduces the performance degradation caused by truncation by 82%-95% compared to the baseline, making prompt-based streaming robust enough for real-world deployment.

Related papers

• Generative Transmission: Rethinking Computation, Bandwidth, and Memory in Communication [50.47339685860709]
  We propose Generative Transmission (GenTrans) for video communication under ultra-low-bandwidth and weak-network conditions.<n>By leveraging generative priors, cross-clip memory reuse, runtime state reuse, and weak-network-aware transport, GenTrans significantly reduces transmission overhead.<n> Experimental results show that GenTrans supports effective video transmission under ultra-low-bitrate and weak-network conditions.
  arXiv  Detail & Related papers  (2026-07-20T02:11:26Z)
• TextOVSR: Text-Guided Real-World Opera Video Super-Resolution [72.569334239113]
  Classic opera videos exhibit poor visual quality due to limitations of early filming equipment and long-term degradation during storage.<n>We propose a Text-guided Dual-Branch Opera Video Super-Resolution network, which introduces two types of textual prompts to guide the super-resolution process.<n> Experiments on our OperaLQ benchmark show that TextOVSR outperforms state-of-the-art methods both qualitatively and quantitatively.
  arXiv  Detail & Related papers  (2026-03-16T11:48:43Z)
• Enhancing Neural Video Compression of Static Scenes with Positive-Incentive Noise [51.028614105626154]
  Static scene videos, such as surveillance feeds and videotelephony streams, constitute a dominant share of storage consumption and network traffic.<n>Traditional standardized codecs and neural video compression (NVC) methods struggle to encode these videos efficiently due to inadequate usage of temporal redundancy and severe distribution gaps between training and test data.<n>We propose to incorporate positive-incentive noise into NVC for static scene videos, where short-term temporal changes are reinterpreted as positive-incentive noise to facilitate model finetuning.
  arXiv  Detail & Related papers  (2026-03-06T09:47:25Z)
• High-Fidelity Causal Video Diffusion Models for Real-Time Ultra-Low-Bitrate Semantic Communication [43.045237067909106]
  We introduce a video diffusion model for high-fidelity, causal, and real-time video generation under ultra-low-bitrate semantic communication constraints.<n>Our framework achieves strong perceptual quality, semantic fidelity, and temporal consistency at ultra-lows ( 0.0003 bpp), outperforming classical, neural, and generative baselines in extensive quantitative, qualitative, and subjective evaluations.
  arXiv  Detail & Related papers  (2026-02-14T17:23:55Z)
• Resi-VidTok: An Efficient and Decomposed Progressive Tokenization Framework for Ultra-Low-Rate and Lightweight Video Transmission [35.3961976297755]
  Resi-VidTok is a Resilient Tokenization-Enabled framework for ultra-low-rate and lightweight video transmission.<n>A key contribution is a resilient 1D tokenization pipeline for video that integrates differential temporal token coding.<n>Results indicate robust visual and semantic consistency at channel bandwidth ratios (CBR) as low as 0.0004 and real-time reconstruction at over 30 fps.
  arXiv  Detail & Related papers  (2025-10-28T22:02:36Z)
• InfiniteTalk: Audio-driven Video Generation for Sparse-Frame Video Dubbing [66.48064661467781]
  We introduce sparse-frame video dubbing, a novel paradigm that strategically preserves references to maintain identity, iconic gestures, and camera trajectories.<n>We propose InfiniteTalk, a streaming audio-driven generator designed for infinite-length long sequence dubbing.<n> Comprehensive evaluations on HDTF, CelebV-HQ, and EMTD datasets demonstrate state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-08-19T17:55:23Z)
• PromptMobile: Efficient Promptus for Low Bandwidth Mobile Video Streaming [18.32444457609669]
  This paper presents PromptMobile, an efficient acceleration framework tailored for on-device Promptus.<n>Compared with other streaming methods, PromptMobile achives an average LPIPS improvement of 0.016 (compared with H.265), reducing 60% of severely distorted frames (compared to VQGAN)
  arXiv  Detail & Related papers  (2025-03-20T13:00:36Z)
• Promptus: Can Prompts Streaming Replace Video Streaming with Stable Diffusion [6.093768205619601]
  We propose Promptus, a disruptive communication system that streaming prompts instead of video content.<n>It represents real-world video frames with a series of "prompts" for delivery and employs Stable Diffusion to generate videos at the receiver.<n>Our work opens up a new paradigm for efficient video communication.
  arXiv  Detail & Related papers  (2024-05-30T13:16:48Z)
• Ultra-low bitrate video conferencing using deep image animation [7.263312285502382]
  We propose a novel deep learning approach for ultra-low video compression for video conferencing applications. We employ deep neural networks to encode motion information as keypoint displacement and reconstruct the video signal at the decoder side.
  arXiv  Detail & Related papers  (2020-12-01T09:06:34Z)
• Multi-level Wavelet-based Generative Adversarial Network for Perceptual Quality Enhancement of Compressed Video [51.631731922593225]
  Existing methods mainly focus on enhancing the objective quality of compressed video while ignoring its perceptual quality. We propose a novel generative adversarial network (GAN) based on multi-level wavelet packet transform (WPT) to enhance the perceptual quality of compressed video.
  arXiv  Detail & Related papers  (2020-08-02T15:01:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.