oneoneliu / ggcnn_cornell_dataset Goto Github PK
View Code? Open in Web Editor NEWcornell grasp dataset analyses and process
cornell grasp dataset analyses and process
Hello, because the Cornell Grasping Dataset link is invalid, do you have any files you can provide to me? Thank you
Could I ask you one question?What is the address where you download this cornell dataset?I can't log on to the previous website.
hello,大佬,我想用您的代码去预测单张图片的抓取点,请问在您的基础上如何修改,感谢!!!
Hello, when comparing your code with that of dougsm https://github.com/dougsm/ggcnn, I found that both of you predict width instead of length.
As you've mentioned in your code, the definition of length and width is different in your code from that of dougsm. Did he make a mistake?This question also bothers me a long time.
朋友你好,我看到了在baxter-pnp你给我的关于配置仿真环境的回复,我现在也没有成功,也是在学习阶段的一个新手,能加个联系方式一起交流吗
你好,我是在学习了您这个代复现的笔记,目前正在学习,我的想法是 想在此基础上出一些视频教程,版权我会注明,不知您是否同意?
角度范围-180到180不可以吗?为什么求正余弦要对角度的两倍求,直接原角度求正余弦再arctan2反求角度不行吗(前提角度限定范围在-180,180)?
还有刚试了试把标签当作网络输出去验证,不知道为什么都不匹配
`
import torch
import torch.optim as optim
import numpy as np
import matplotlib.pyplot as plt
import cv2
import time
#导入自定义包
from ggcnn import GGCNN
from cornell_pro import Cornell
from functions import post_process,detect_grasps,max_iou
from image_pro import Image
from grasp_pro import Grasp_cpaw
val_data = Cornell('图片标签路径',
random_rotate = True,
random_zoom = True,
output_size = 300)
val_dataset = torch.utils.data.DataLoader(val_data,batch_size = 1,shuffle = True)
for x,y,idx,rot,zoom in val_dataset:
q_out,ang_out,width_out = post_process(y[0], y[1], y[2], y[3])
grasps_pre = detect_grasps(q_out,ang_out,width_out,no_grasp = 1)
grasps_true = val_dataset.dataset.get_raw_grasps(idx,rot,zoom)
result = 0
for grasp_pre in grasps_pre:
if max_iou(grasp_pre,grasps_true) > 0.25:
result = 1
break
if result==0:
print(val_dataset.dataset.graspf[idx])
#break
`
Hello, I am confused by the aixs X and Y in the function as_gr.
def as_gr(self):
"""
Convert to GraspRectangle
:return: GraspRectangle representation of grasp.
"""
xo = np.cos(self.angle)
yo = np.sin(self.angle)
y1 = self.center[0] + self.length / 2 * yo
x1 = self.center[1] - self.length / 2 * xo
y2 = self.center[0] - self.length / 2 * yo
x2 = self.center[1] + self.length / 2 * xo
return GraspRectangle(np.array(
[
[y1 - self.width/2 * xo, x1 - self.width/2 * yo],
[y2 - self.width/2 * xo, x2 - self.width/2 * yo],
[y2 + self.width/2 * xo, x2 + self.width/2 * yo],
[y1 + self.width/2 * xo, x1 + self.width/2 * yo],
]
).astype(np.float))
The definition of axis X and Y is different for images and in mathsmatics. Does the code use the definition in mathsmatics instead of in imgae in the code above?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.