[enoch2090]

Surprisingly, SAM3 works bad on engineering drawings while SAM2 kinda works, and VLMs like Qwen3-VL works as well

[zubiaur]

Had good luck with Gemini 2.5, SAM3 failed miserably with PIDs.

[retinaros]

yeah I tried too. Im trying a fine tuning on PIDs.

[enoch2090]

Looking forward to your progress! Just checked the paper and it says the underlying backbone is still DETR. My guess would be that SAM3 uses more video frames during the training process and caused the dilution of sparse engineering-paper-like data.