jzx
``` process[ssl_slam2_map_optimization_node-3]: started with pid [29347] double free or corruption (out) process[baselink2cam_tf-4]: started with pid [29391] process[rviz-5]: started with pid [29393] process[ssl_slam2/trajectory_server_ssl_slam-6]: started with pid [29398] [ssl_slam2_map_optimization_node-3] process has died...
感谢作者，这是一项很赞的工作！！！我想从detect everything的角度上来使用这个模型，所以选取了同一张测试图片，与detic做了推理上的比较,有两个问题： 1.yolo-world的检测confidence会稍微有些低，但这不影响使用 2.yolo-world在这张图片上检出率会稍微差一些，是因为detic是大模型的缘故？  and i use 'cone' world set confidence to 0.01   
nice work！ I have deploy my detection model in tensorrt and it could infer vector,i have 8 rtsp camera,i use opencv to get the 8 rtsp camera and make the...
RT In my opinion, bounding box detection is simpler and lighter than semantic segmentation
Awesome work! I want to segment anything ,get a full segment mask From an image using Nano Sam like deeplabv3 What should I do?thank you very much Sent from PPHub
Similar to mild, severe, pink, deformed, short, etc like ：a tall man and a short man is two class Thank you very much！
the diffusion is very nice：https://github.com/real-stanford/diffusion_policy Can xarm robot support this repo as hardware.it will be very cool and popular!