AI 精选动态
智能评分 67
C-Fast-FoundationStereo 发布
AI 推荐理由
与 FoundationStereo 相比，该模型通过蒸馏和剪枝实现 10 倍加速且精度几乎无损，值得关注其技术细节和开源代码。核心解读
NVIDIA 发布 C-Fast-FoundationStereo，基于 FoundationStereo 进行蒸馏、架构搜索和结构化剪枝，参数量 14.6M，在零样本立体深度估计上比 FoundationStereo 快 10 倍以上且精度接近。支持 PyTorch、NVIDIA TAO、TensorRT 和 ONNXRuntime 导出。
全文
Real-time stereo depth, with FoundationStereo-style zero-shot generalization. ⚡
C-Fast-FoundationStereo is now on ModelScope!
🔗 https://t.co/1M2N1FnaLG
Key points:
🚀 Over 10× faster than FoundationStereo
🎯 Closely matches its zero-shot accuracy
👁️ Input: rectified left-right RGB stereo images
🗺️ Output: dense disparity map for depth estimation
🧠 14.6M parameters
🛠️ Built with distillation, architecture search, and structured pruning
Supports PyTorch, NVIDIA TAO, TensorRT, and ONNXRuntime export.
License: NVIDIA Open Model Agreement
