🧨EfficientFormers: 1.6ms inference 🧨

👉Transformers fast as MobileNet? Snap shows that on #iphone!

𝐇𝐢𝐠𝐡𝐥𝐢𝐠𝐡𝐭𝐬:
✅Low latency on mobile, high performance!
✅Revisiting the design of ViT through latency
✅New dimension-consistent design paradigm
✅EfficientFormers: a new ViT for mobile!

More: https://bit.ly/3MdgW15

📲 EdgeSAM: Mobile 40x SAM 📲

👉A novel hyper-optimized version of SAM for mobile devices such as #Iphone. Pure CNNs backbone (better suitable for ANE), up to 40x faster. Code available 😉

👉Review https://t.ly/m_vLH
👉Paper https://lnkd.in/gHZVZN2x
👉Project https://lnkd.in/gK8qEK8p
👉Repo https://lnkd.in/gj6YAGNv
👉Hugging Face https://lnkd.in/gUUHJvxz