SenseTime@SenseTime_AI

2026-07-13 22:41· 8天前

AI 摘要

商汤开源 SenseNova-Vision-7B-MoT，该模型将计算机视觉重构为多模态生成，无需任务专用头即可达到 SOTA，在分割与密集几何任务上超越 Google DeepMind 的 Vision Banana。具体对比：Referring segmentation（RefCOCOg cIoU）80.3 vs 73.8，Semantic segmentation（Cityscapes mIoU）71.2 vs 69.9，Depth estimation（NYUv2 δ1）98.1 vs 94.8，Surface normal（NYUv2 mean error）14.4 vs 17.8。模型支持语言或视觉提示引导，基于商汤十年 CV 积累。新任务由指令定义，而非新模型头。

Open-sourcing SenseNova-Vision-7B-MoT - one model， all major vision tasks.

Reimagines CV as multimodal generation. Steerable by language or visual prompts. No task-specific heads - still SOTA， surpassing Google DeepMind's Vision Banana on segmentation and dense geometry.

📊 vs. Vision Banana （Google DeepMind）： 🔹Referring segmentation （RefCOCOg cIoU）： 80.3 vs 73.8 🔹Semantic segmentation （Cityscapes mIoU）： 71.2 vs 69.9 🔹Depth estimation （NYUv2 δ1）： 98.1 vs 94.8 🔹Surface normal （NYUv2 mean error）： 14.4 vs 17.8

Built on a decade of SenseTime's CV leadership - #1 in China's Vision AI market for 10 consecutive years， and named a Tech Innovator in global GenAI CV by Gartner. Vision， now native to foundation models. New tasks， defined by instruction - not by new model heads.

多模态开源生态模型发布

SenseTime@SenseTime_AI · X

55导出 Markdown

2026-07-13 22:41·8天前

在 X 看原推· x.com

AI 摘要

Open-sourcing SenseNova-Vision-7B-MoT - one model， all major vision tasks.

Reimagines CV as multimodal generation. Steerable by language or visual prompts. No task-specific heads - still SOTA， surpassing Google DeepMind's Vision Banana on segmentation and dense geometry.