🚀 Do check out our latest work on Modular Mixed-precision Quantization!

🌐 Page: saqibjaved1.github.io/MQAT_/

#Quantization #Mixedprecision #DeepLearning #Robotics #AI #6Dpose #edgeAI

🚀 Excited to share our TMLR publication on Modular Quantization-Aware Training (MQAT)!
We show how modular mixed-precision quantization boosts performance & reduces model size for edge devices! 🎯

🌐 Page: saqibjaved1.github.io/MQAT_/

#DeepLearning #Robotics #AI #6Dpose #edgeAI

A new single-reference 6D pose estimation method reaches CAD-level accuracy, matching traditional multi-view precision while using only one reference image. Read more: https://getnews.me/single-reference-6d-pose-estimation-achieves-cad-level-accuracy/ #6dpose #computervision #cadaccuracy

Kuanning Wang, Yuqian Fu, Tianyu Wang, Yanwei Fu, Longfei Liang, Yu-Gang Jiang, Xiangyang Xue: RAG-6DPose: Retrieval-Augmented 6D Pose Estimation via Leveraging CAD as Knowledge Base https://arxiv.org/abs/2506.18856 https://arxiv.org/pdf/2506.18856 https://arxiv.org/html/2506.18856

Researchers unveiled a two-stage framework using a lighting‑invariant color‑pair descriptor for 6D pose estimation, achieving accuracy comparable to state‑of‑the‑art methods on edge devices. https://getnews.me/robust-zero-shot-6d-pose-estimation-and-tracking-for-edge-devices/ #6dpose #edgedevice

SingRef6D estimates 6D pose from a single RGB image, removing depth sensors. It improves the δ1.05 metric by 14.41% on REAL275 and raises average recall by 6.1%. Read more: https://getnews.me/singref6d-single-image-6d-pose-estimation-without-depth-sensors/ #singref6d #6dpose

Kuanning Wang, Yuqian Fu, Tianyu Wang, Yanwei Fu, Longfei Liang, Yu-Gang Jiang, Xiangyang Xue
RAG-6DPose: Retrieval-Augmented 6D Pose Estimation via Leveraging CAD as Knowledge Base
https://arxiv.org/abs/2506.18856