SuperPoint学习训练纪录 无训练版与带训练版本（一）
2024-05-15 16:08:36
superpoint 论文原文，先过一遍论文。
superpoint 官方源代码，基于 pytorch 的，不过遗憾的是训练代码和相应的渲染的训练数据没提供。
superpoint 大佬复现带训练版本，基于tensorflow。
1.官方源码效果
命令行下载源码，也可以点上面链接下载。
git clone https://github.com/magicleap/SuperPointPretrainedNetwork.git 先配置环境（强烈建议安装anaconda配置虚拟环境，方便管理）
pip install opencv-python
pip install torch
我的实现版本：opencv-python 4.4 torch1.8.1
官方要求：
运行图片模式（其他模式github都有对应命令代码，输入为摄像头或视频）
./demo_superpoint.py assets/给到图片文件夹的路径上一级/
注意图片不是png格式则需要做修改，在img_glob default里修改，运行卡顿也可以改HW调整输入图片大小。
跟踪效果（检测对应特征点在图片上的移动）
源码只有跟踪效果，我在源码的基础上做了一些修改，写了匹配和用时匹配数等数据的可视化。
import argparse
import glob
import numpy as np
import os
import time
import cv2
from numpy.core.records import array
from numpy.distutils.system_info import x11_info
import torch
if int(cv2.__version__[0]) < 3: # pragma: no coverprint('Warning: OpenCV 3 is not installed')
class SuperPointNet(torch.nn.Module):#def __init__(self):super(SuperPointNet, self).__init__() # 第一句话，调用父类的构造函数 也就是nn.moduleself.relu = torch.nn.ReLU(inplace=True)self.pool = torch.nn.MaxPool2d(kernel_size=2, stride=2)c1, c2, c3, c4, c5, d1 = 64, 64, 128, 128, 256, 256# Shared Encoder.self.conv1a = torch.nn.Conv2d(1, c1, kernel_size=3, stride=1, padding=1)# 编码卷集层self.conv1b = torch.nn.Conv2d(c1, c1, kernel_size=3, stride=1, padding=1)self.conv2a = torch.nn.Conv2d(c1, c2, kernel_size=3, stride=1, padding=1)self.conv2b = torch.nn.Conv2d(c2, c2, kernel_size=3, stride=1, padding=1)self.conv3a = torch.nn.Conv2d(c2, c3, kernel_size=3, stride=1, padding=1)self.conv3b = torch.nn.Conv2d(c3, c3, kernel_size=3, stride=1, padding=1)self.conv4a = torch.nn.Conv2d(c3, c4, kernel_size=3, stride=1, padding=1)self.conv4b = torch.nn.Conv2d(c4, c4, kernel_size=3, stride=1, padding=1)# Detector Head.self.convPa = torch.nn.Conv2d(c4, c5, kernel_size=3, stride=1, padding=1)# 解码卷集层self.convPb = torch.nn.Conv2d(c5, 65, kernel_size=1, stride=1, padding=0)# Descriptor Head.self.convDa = torch.nn.Conv2d(c4, c5, kernel_size=3, stride=1, padding=1)# 解码卷集层self.convDb = torch.nn.Conv2d(c5, d1, kernel_size=1, stride=1, padding=0)def forward(self, x):# Shared Encoder.x = self.relu(self.conv1a(x))x = self.relu(self.conv1b(x))x = self.pool(x)x = self.relu(self.conv2a(x))x = self.relu(self.conv2b(x))x = self.pool(x)x = self.relu(self.conv3a(x))x = self.relu(self.conv3b(x))x = self.pool(x)x = self.relu(self.conv4a(x))x = self.relu(self.conv4b(x))# Detector Head.cPa = self.relu(self.convPa(x))semi = self.convPb(cPa)# Descriptor Head.cDa = self.relu(self.convDa(x))desc = self.convDb(cDa)dn = torch.norm(desc, p=2, dim=1) # Compute the norm.desc = desc.div(torch.unsqueeze(dn, 1)) # Divide by norm to normalize.return semi, desc #网络结构 返回值class SuperPointFrontend(object): #
#superpoint前端类 继承了object
#包含住了superpoint类 """ Wrapper around pytorch net to help with pre and post image processing. """def __init__(self, weights_path, nms_dist, conf_thresh, nn_thresh,cuda=False):self.name = 'SuperPoint'self.cuda = cudaself.nms_dist = nms_distself.conf_thresh = conf_thresh#传参self.nn_thresh = nn_thresh # L2 descriptor distance for good match.self.cell = 8 # Size of each output cell. Keep this fixed.每个输出单元格的大小。保持固定的。self.border_remove = 4 # Remove points this close to the border. 边界点# Load the network in inference mode.self.net = SuperPointNet() #子类if cuda:# Train on GPU, deploy on GPU.self.net.load_state_dict(torch.load(weights_path))self.net = self.net.cuda()else:# Train on GPU, deploy on CPU.self.net.load_state_dict(torch.load(weights_path,map_location=lambda storage, loc: storage))self.net.eval()def nms_fast(self, in_corners, H, W, dist_thresh):#非极大值抑制 ，dis thresh 默认4grid = np.zeros((H, W)).astype(int) # Track NMS data. #array inds = np.zeros((H, W)).astype(int) # 存储点的索引。# 根据特征点信心值排序 四舍五入最接近的int。inds1 = np.argsort(-in_corners[2,:])#argsort返回数组值从小到大的索引值 ，前是行索引 后面是列 取第三行的信心值全切片 从大到小排序corners = in_corners[:,inds1]#取出所有特征点信值rcorners = corners[:2,:].round().astype(int) # 取出特征点xy 0-2 列全部# Check for edge case of 0 or 1 corners.检查0或1个角的边缘情况。if rcorners.shape[1] == 0:return np.zeros((3,0)).astype(int), np.zeros(0).astype(int)if rcorners.shape[1] == 1:out = np.vstack((rcorners, in_corners[2])).reshape(3,1)return out, np.zeros((1)).astype(int)# Initialize the grid.for i, rc in enumerate(rcorners.T):#enumerate就是枚举的意思，把元素一个个列举出来, 所以他返回的是元素rc以及对应的索引i。grid[rcorners[1,i], rcorners[0,i]] = 1 #有哪些点 有就给1inds[rcorners[1,i], rcorners[0,i]] = i #给点对应的特征点序号# Pad the border of the grid, so that we can NMS points near the border.pad = dist_thresh #距离4 intgrid = np.pad(grid, ((pad,pad), (pad,pad)), mode='constant')#pad填充边缘网格 constant default=0# Iterate through points, highest to lowest conf, suppress neighborhood.#遍历点，从最高到最低的conf，抑制邻域。count = 0for i, rc in enumerate(rcorners.T):# Account for top and left padding.pt = (rc[0]+pad, rc[1]+pad)#4*4领域 因为要考虑边界if grid[pt[1], pt[0]] == 1: # If not yet suppressed.grid[pt[1]-pad:pt[1]+pad+1, pt[0]-pad:pt[0]+pad+1] = 0grid[pt[1], pt[0]] = -1 #抑制完了取-1count += 1# Get all surviving -1's and return sorted array of remaining corners.#获取所有幸存的-1，并返回剩余角的排序数组。keepy, keepx = np.where(grid==-1) #-1的xy where返回坐标keepy, keepx = keepy - pad, keepx - pad#真实的xy坐标 因为前面加4了inds_keep = inds[keepy, keepx] out = corners[:, inds_keep] #取出xy和confvalues = out[-1, :] #出confinds2 = np.argsort(-values) #排序conf 返回索引out = out[:, inds2]out_inds = inds1[inds_keep[inds2]]return out, out_inds#剩下点的3*n 和索引def run(self, img):assert img.ndim == 2, 'Image must be grayscale.'assert img.dtype == np.float32, 'Image must be float32.'H, W = img.shape[0], img.shape[1]inp = img.copy()inp = (inp.reshape(1, H, W))inp = torch.from_numpy(inp)inp = torch.autograd.Variable(inp).view(1, 1, H, W)if self.cuda:inp = inp.cuda()# Forward pass of network.outs = self.net.forward(inp)semi, coarse_desc = outs[0], outs[1]# Convert pytorch -> numpy.semi = semi.data.cpu().numpy().squeeze()# --- Process points.dense = np.exp(semi) # Softmax.dense = dense / (np.sum(dense, axis=0)+.00001) # Should sum to 1.# Remove dustbin.nodust = dense[:-1, :, :]# Reshape to get full resolution heatmap.Hc = int(H / self.cell)Wc = int(W / self.cell)nodust = nodust.transpose(1, 2, 0)heatmap = np.reshape(nodust, [Hc, Wc, self.cell, self.cell])heatmap = np.transpose(heatmap, [0, 2, 1, 3])heatmap = np.reshape(heatmap, [Hc*self.cell, Wc*self.cell])xs, ys = np.where(heatmap >= self.conf_thresh) # Confidence threshold.if len(xs) == 0:return np.zeros((3, 0)), None, Nonepts = np.zeros((3, len(xs))) # Populate point data sized 3xN.pts[0, :] = yspts[1, :] = xspts[2, :] = heatmap[xs, ys]pts, _ = self.nms_fast(pts, H, W, dist_thresh=self.nms_dist) # Apply NMS.inds = np.argsort(pts[2,:])pts = pts[:,inds[::-1]] # Sort by confidence.# Remove points along border.bord = self.border_removetoremoveW = np.logical_or(pts[0, :] < bord, pts[0, :] >= (W-bord))toremoveH = np.logical_or(pts[1, :] < bord, pts[1, :] >= (H-bord))toremove = np.logical_or(toremoveW, toremoveH)pts = pts[:, ~toremove]# --- Process descriptor.D = coarse_desc.shape[1]if pts.shape[1] == 0:desc = np.zeros((D, 0))else:# Interpolate into descriptor map using 2D point locations.samp_pts = torch.from_numpy(pts[:2, :].copy())samp_pts[0, :] = (samp_pts[0, :] / (float(W)/2.)) - 1.samp_pts[1, :] = (samp_pts[1, :] / (float(H)/2.)) - 1.samp_pts = samp_pts.transpose(0, 1).contiguous()samp_pts = samp_pts.view(1, 1, -1, 2)samp_pts = samp_pts.float()if self.cuda:samp_pts = samp_pts.cuda()desc = torch.nn.functional.grid_sample(coarse_desc, samp_pts)desc = desc.data.cpu().numpy().reshape(D, -1)desc /= np.linalg.norm(desc, axis=0)[np.newaxis, :]return pts, desc, heatmapdef nn_match_two_way(desc1, desc2, nn_thresh):assert desc1.shape[0] == desc2.shape[0]if desc1.shape[1] == 0 or desc2.shape[1] == 0:return np.zeros((3, 0))if nn_thresh < 0.0:raise ValueError('\'nn_thresh\' should be non-negative')# Compute L2 distance. Easy since vectors are unit normalized.dmat = np.dot(desc1.T, desc2)dmat = np.sqrt(2-2*np.clip(dmat, -1, 1))# Get NN indices and scores.idx = np.argmin(dmat, axis=1)scores = dmat[np.arange(dmat.shape[0]), idx]# Threshold the NN matches.keep = scores < nn_thresh# Check if nearest neighbor goes both directions and keep those.idx2 = np.argmin(dmat, axis=0)keep_bi = np.arange(len(idx)) == idx2[idx]keep = np.logical_and(keep, keep_bi)idx = idx[keep]scores = scores[keep]# Get the surviving point indices.m_idx1 = np.arange(desc1.shape[1])[keep]m_idx2 = idx# Populate the final 3xN match data structure.matches = np.zeros((3, int(keep.sum())))matches[0, :] = m_idx1matches[1, :] = m_idx2matches[2, :] = scoresreturn matches class VideoStreamer(object):def __init__(self, basedir, camid, height, width, skip, img_glob):#img_glod *.png # 构造函数# 图模式中主要用来得到self.list 所有图片的路径#self.cap = [] #listself.camera = Falseself.video_file = Falseself.listing = []self.sizer = [height, width]self.i = 0self.skip = skipself.maxlen = 1000000# If the "basedir" string is the word camera, then use a webcam.if basedir == "camera/" or basedir == "camera": #使用相机print('==> Processing Webcam Input.')self.cap = cv2.VideoCapture(camid)self.listing = range(0, self.maxlen)self.camera = Trueelse:# Try to open as a video.self.cap = cv2.VideoCapture(basedir) #使用视屏lastbit = basedir[-4:len(basedir)] #len str的长度 string[-4:最后] 切片 切最后四个字符if (type(self.cap) == list or not self.cap.isOpened()) and (lastbit == '.mp4'):raise IOError('Cannot open movie file')elif type(self.cap) != list and self.cap.isOpened() and (lastbit != '.txt'):print('==> Processing Video Input.')num_frames = int(self.cap.get(cv2.CAP_PROP_FRAME_COUNT)) #视频总帧数self.listing = range(0, num_frames) #生成listself.listing = self.listing[::self.skip] #切片 0-num_frame 10步一切self.camera = Trueself.video_file = Trueself.maxlen = len(self.listing) #更新总帧数else:print('==> Processing Image Directory Input.')search = os.path.join(basedir, img_glob)#前后str拼起来 得到*图片路径 名字用*.png#type(search)=strself.listing = glob.glob(search) #找到所有符合search格式的图片路径self.listing.sort() #从小到大排序self.listing = self.listing[::self.skip] ##切片 遍历list skip步一切self.maxlen = len(self.listing) #更新总帧数if self.maxlen == 0:raise IOError('No images were found (maybe bad \'--img_glob\' parameter?)')def read_image(self, impath, img_size): #读图的成员函数""" Read image as grayscale and resize to img_size.Inputsimpath: Path to input image.img_size: (W, H) tuple specifying resize size.Returnsgrayim: float32 numpy array sized H x W with values in range [0, 1]."""grayim = cv2.imread(impath, 0) #0 grayif grayim is None:raise Exception('Error reading image %s' % impath)# Image is resized via opencv.interp = cv2.INTER_AREA#一种插值方法 一般用来缩小grayim = cv2.resize(grayim, (img_size[1], img_size[0]), interpolation=interp) #把图片缩小到指定HWgrayim = (grayim.astype('float32') / 255.) #格式转换return grayimdef next_frame(self): #依次给图的成员函数""" Return the next frame, and increment internal counter.Returnsimage: Next H x W image.status: True or False depending whether image was loaded."""if self.i == self.maxlen:#到最后一帧了return (None, False)if self.camera: #相机模式ret, input_image = self.cap.read()if ret is False:print('VideoStreamer: Cannot get image from camera (maybe bad --camid?)')return (None, False)if self.video_file: #video模式self.cap.set(cv2.CAP_PROP_POS_FRAMES, self.listing[self.i])input_image = cv2.resize(input_image, (self.sizer[1], self.sizer[0]),interpolation=cv2.INTER_AREA)input_image = cv2.cvtColor(input_image, cv2.COLOR_RGB2GRAY)input_image = input_image.astype('float')/255.0else:image_file = self.listing[self.i] #切当前片input_image = self.read_image(image_file, self.sizer)# Increment internal counter.self.i = self.i + 1#内部的计数器input_image = input_image.astype('float32')return (input_image, True)def match_descriptors(kp1, desc1, kp2, desc2):# Match the keypoints with the warped_keypoints with nearest neighbor searchbf = cv2.BFMatcher(cv2.NORM_L2, crossCheck=True)##是一个暴力匹配的对象，取desc1中一个描述子，再与desc2中的所有描述子计算欧式距离matches = bf.match(desc1, desc2)##上一行的返回值类似元组的集合（i，j）代表第一个集合的第i个点的最佳匹配是第二个集合的第j个点matches_idx = np.array([m.queryIdx for m in matches]) m_kp1 = [kp1[idx] for idx in matches_idx]matches_idx = np.array([m.trainIdx for m in matches])m_kp2 = [kp2[idx] for idx in matches_idx]####m_kp1是第一张图片的特征点，m_kp2是第二张图片的特征点，此时它们已经一一对应了（至少是最对应的，距离最小的return m_kp1, m_kp2, matches def showpoint(img,ptx):for i in range(ptx.shape[1]):x=int(round(ptx[0,i]))y=int(round(ptx[1,i]))# if x>20 and y>20 and x<640 and y <450:# Nonecv2.circle(img,(x,y),3,color=(255,0,0))return imgdef drawMatches(img1, kp1, img2, kp2, matches):"""My own implementation of cv2.drawMatches as OpenCV 2.4.9does not have this function available but it's supported inOpenCV 3.0.0This function takes in two images with their associated keypoints, as well as a list of DMatch data structure (matches) that contains which keypoints matched in which images.An image will be produced where a montage is shown withthe first image followed by the second image beside it.Keypoints are delineated with circles, while lines are connectedbetween matching keypoints.img1,img2 - Grayscale imageskp1,kp2 - Detected list of keypoints through any of the OpenCV keypoint detection algorithmsmatches - A list of matches of corresponding keypoints through anyOpenCV keypoint matching algorithm"""# Create a new output image that concatenates the two images together# (a.k.a) a montagerows1 = img1.shape[0]cols1 = img1.shape[1]rows2 = img2.shape[0]cols2 = img2.shape[1]#out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')# Place the first image to the lefti1= np.dstack([img1, img1, img1])i2=np.dstack([img2, img2, img2])cv2.imshow("sdd",i1)cv2.imshow("sd",i1)out = np.hstack([i1,i2])print("sdsdsd",out.shape)# Place the next image to the right of it#out[0:480,640:1280] = np.dstack([img2, img2, img2])# For each pair of points we have between both images# draw circles, then connect a line between themfor i in range(matches.shape[1]):# Get the matching keypoints for each of the imagesimg1_idx = matches[0,i]img2_idx = matches[1,i]x11=int(img1_idx)y11=int(img1_idx)x22=int(img2_idx)y22=int(img2_idx)# x - columns# y - rowsx1= kp1[0,x11]y1= kp1[1,y11]x2= kp2[0,x22]y2= kp2[1,y22]# Draw a small circle at both co-ordinates# radius 4# colour blue# thickness = 1a = np.random.randint(0,256)b = np.random.randint(0,256)c = np.random.randint(0,256)cv2.circle(out, (int(np.round(x1)),int(np.round(y1))), 2, (a, b, c), 1) #画圆，cv2.circle()参考官方文档cv2.circle(out, (int(np.round(x2)+cols1),int(np.round(y2))), 2, (a, b, c), 1)# Draw a line in between the two points# thickness = 1# colour bluecv2.line(out, (int(np.round(x1)),int(np.round(y1))), (int(np.round(x2)+cols1),int(np.round(y2))), (a, b, c), 1, shift=0) #画线，cv2.line()参考官方文档# Also return the image if you'd like a copyreturn outif __name__ == '__main__':# Parse command line arguments.parser = argparse.ArgumentParser(description='PyTorch SuperPoint Demo.')parser.add_argument('input', type=str, default='',help='Image directory or movie file or "camera" (for webcam).')parser.add_argument('--weights_path', type=str, default='superpoint_v1.pth',help='Path to pretrained weights file (default: superpoint_v1.pth).')parser.add_argument('--img_glob', type=str, default='*.png',#################pgmhelp='Glob match if directory of images is specified (default: \'*.png\').')parser.add_argument('--skip', type=int, default=1,help='Images to skip if input is movie or directory (default: 1).')parser.add_argument('--show_extra', action='store_true',help='Show extra debug outputs (default: False).')parser.add_argument('--H', type=int, default=480,help='Input image height (default: 120).')parser.add_argument('--W', type=int, default=640,help='Input image width (default:640).')parser.add_argument('--display_scale', type=int, default=2,help='Factor to scale output visualization (default: 2).')parser.add_argument('--min_length', type=int, default=2,help='Minimum length of point tracks (default: 2).')parser.add_argument('--max_length', type=int, default=5,help='Maximum length of point tracks (default: 5).')parser.add_argument('--nms_dist', type=int, default=4,help='Non Maximum Suppression (NMS) distance (default: 4).')parser.add_argument('--conf_thresh', type=float, default=0.015,help='Detector confidence threshold (default: 0.015).')parser.add_argument('--nn_thresh', type=float, default=0.7,help='Descriptor matching threshold (default: 0.7).')parser.add_argument('--camid', type=int, default=0,help='OpenCV webcam video capture ID, usually 0 or 1 (default: 0).')parser.add_argument('--waitkey', type=int, default=1,help='OpenCV waitkey time in ms (default: 1).')parser.add_argument('--cuda', action='store_true',help='Use cuda GPU to speed up network processing speed (default: False)')parser.add_argument('--no_display', action='store_true',help='Do not display images to screen. Useful if running remotely (default: False).')parser.add_argument('--write', action='store_true',help='Save output frames to a directory (default: False)')parser.add_argument('--write_dir', type=str, default='tracker_outputs/',help='Directory where to write output frames (default: tracker_outputs/).')opt = parser.parse_args()print(opt)#读图 读下一张图vs = VideoStreamer(opt.input, opt.camid, opt.H, opt.W, opt.skip, opt.img_glob)print('==> Loading pre-trained network.')# This class runs the SuperPoint network and processes its outputs.fe = SuperPointFrontend(weights_path=opt.weights_path,#权重的路径strnms_dist=opt.nms_dist,#非极大值抑制 int距离4conf_thresh=opt.conf_thresh,#探测器阈值0.015nn_thresh=opt.nn_thresh,#匹配器阈值0.7cuda=opt.cuda) #GPU加速 默认falseprint('==> Successfully loaded pre-trained network.')# Create a window to display the demo.if not opt.no_display:win = 'SuperPoint Tracker'cv2.namedWindow(win)else:print('Skipping visualization, will not show a GUI.')# Font parameters for visualizaton.font = cv2.FONT_HERSHEY_DUPLEX#设置可视化字体font_clr = (255, 255, 255)font_pt = (4, 12)font_sc = 0.4#创建输出目录if opt.write:#默认falseprint('==> Will write outputs to %s' % opt.write_dir)if not os.path.exists(opt.write_dir):os.makedirs(opt.write_dir)print('==> Running Demo.')img1, status = vs.next_frame()#读第一张图start1 = time.time()pts, desc, heatmap = fe.run(img1)end1 = time.time()c2=end1-start1print("第一张图提取用时",c2,"提取特征点数目",pts.shape[1])imgx=img1.copy()img11=showpoint(imgx,pts)cv2.imshow("imgone",img11)img2, status = vs.next_frame()#读第二张图start1 = time.time()pts1, desc1, heatmap1 = fe.run(img2)end1 = time.time()c2=end1-start1print("第二张图提取用时",c2,"提取特征点数目",pts1.shape[1])imgx=img2.copy()img22=showpoint(imgx,pts1)cv2.imshow("imgtwo",img22)match=nn_match_two_way(desc,desc1,0.7)print("图1与图2匹配对数",match.shape[1])# cv_kpts1 = [cv2.KeyPoint(pts[0,i], pts[1,i], 1)
# for i in range(pts.shape[1])]
# cv_kpts2 = [cv2.KeyPoint(pts1[0,i], pts1[1,i], 1)
# for i in range(pts1.shape[1])]
# sift_matched_img = cv2.drawMatches(img1, cv_kpts1, img2,
# cv_kpts2, matches, None,
# matchColor=(0, 255, 0),
# singlePointColor=(0, 0, 255))#手写匹配 有些问题out=sift_matched_img = drawMatches(img1, pts, img2,pts1, match)cv2.namedWindow("matcher",0)cv2.imshow("matcher",out)cv2.waitKey(0)print('==> Finshed Demo.')
代码运行效果
SuperPoint学习训练纪录 无训练版与带训练版本（一）相关推荐
- SuperPoint学习训练纪录 盘点各种报错解决方法（二）
之前介绍了原版SuperPoint,可惜原版没有公开训练代码,接下来看一个复现版本. 为了成功运行这个版本,反复配置无数遍,建议大家就跟着我的方法来,一次就成功. 运行成功,cuda cudnn te ...
- 花书+吴恩达深度学习（二十）构建模型策略（超参数调试、监督预训练、无监督预训练）
目录 0. 前言 1. 学习率衰减 2. 调参策略 3. 贪心监督预训练 4. 贪心逐层无监督预训练 如果这篇文章对你有一点小小的帮助,请给个关注,点个赞喔~我会非常开心的~ 花书+吴恩达深度学习(十 ...
- 本周AI热点回顾：「时空版」Transformer训练速度远超3D CNN；拒绝内卷的AI狼火了！不想抓羊只想躺！...
点击左上方蓝字关注我们 01 「时空版」Transformer训练速度远超3D CNN,提速3倍! Facebook AI推出了全新的视频理解架构TimeSform ...
- 深度学习实战（七）——目标检测API训练自己的数据集（R-FCN数据集制作+训练+测试）
TensorFlow提供的网络结构的预训练权重:https://cloud.tencent.com/developer/article/1006123 将voc数据集转换成.tfrecord格式供te ...
- 【深度学习 学习率，优化器】——深刻解读训练网络时各种学习率，优化器的区别，learning rate, Momentum
机梯度下降及各种更新方法 普通更新 最简单的更新形式是沿着负梯度方向改变参数(因为梯度指向的是上升方向,但是我们通常希望最小化损失函数).假设有一个参数向量x及其梯度dx,那么最简单的更新的形式是: ...
- 基于元学习和自我监督的图像翻译预训练
提示:系读文章的笔记,侵删. 文章目录 前言 一.摘要 二.结论与展望 1.结论:新的小样本多任务图像到图像的转换 2.展望:提高性能和稳定性 三.介绍 四. 背景和相关工作 1 数据集(风暴事件图像 ...
- 少儿全学科知识学习、复习、训练的免费福利，少儿知识训练大师 5.0 全新发布：真人语音、插件题库、全新功能
孩子全学科复习.强化训练学习的必备工具,各种学科和知识问题库无限发挥.任意组合,配合本地.在线语音合成和识别引擎,为您打造真人级朗读和交互效果,更适合少儿的生动学习场景,本软件还是免费的哦! 新版软件 ...
- RCNN学习笔记——第三篇: 实现FRCNN网络训练、评价与预测(附全部源码)
RCNN学习笔记--第三篇: 实现FRCNN网络训练.评价与预测(附全部源码) 本文是个人根据B站大佬Bubbliiiing的FRCNN系列视频同步完成FRCNN训练,记录心得和遇见的问题. 关于RC ...
- 基于图像的三维物体重建：在深度学习时代的最新技术和趋势综述之训练
点击上方"3D视觉工坊",选择"星标" 干货第一时间送达 1利用其他线索 前面几节讨论了直接从二维观测重建三维对象的方法.本节展示了如何使用附加提示,如中间表示 ...
最新文章
- 【论文阅读】Clustering by Shift
- js修改css样式属性_这个笔记《CSS样式的常见属性及值》，让菜鸟轻松学会包粽子...
- ECS 按量付费VPC实例停机不收费FAQ
- 软件设计师 - 系统测试
- Android -- 图片画画板（canvas、paint、bitmap）
- 【Python】PyMySQL 连接 MySQL数据库
- [动态规划] leetcode 10 正则表达式匹配
- 记一次 AXI -id debug
- Opencv step by step - 图像变换
- 洛谷 P2437 蜜蜂路线
- opencv图像连通区域分析
- Atitit..组件化事件化的编程模型--(2)---------Web datagridview 服务器端控件的实现原理and总结
- 注册使用GAC【转载】
- Fiddler抓包快速入门-windows网页抓包
- C语言制作科学计算器
- 公有云、私有云及混合云应用场景辨析
- ubuntu compiz 不能启动
- postgres查询序列_PostgreSQL之序列(Sequence)
- 充电电池的充放电电流-0.2C、1C、2C的含义
- git上传到阿里云code