我尝试从多个图像进行3D重建。我目前正在使用视觉SFM管道，它将Lowe SIFT功能作为.feat和.mat格式。两者都是二进制的，所以我无法用编辑器阅读它们。由于视觉sfm的文档：
使用您自己的功能匹配 1.编写包含所有功能匹配的txt文件 2.将图像（带功能）加载到VisualSFM中 3.使用＆＃34; SfM-＆gt;成对匹配 - ＆gt;导入特征匹配＆＃34; 4.您可以再次使用相同的方法添加功能匹配。
*。这假设您已经检测到这些功能。 *。命令行的相同功能是选项＆＃34; + import＆＃34; *。不要使用＆＃34; SfM-＆gt;成对匹配 - ＆gt;计算缺失匹配＆＃34;，它完全匹配
匹配文件采用以下格式匹配文件=图像匹配=＆lt;匹配数＆gt;
例如，下面给出了888.jpg和之间的24场比赛 709.jpg
** 888.jpg 709.jpg 24
19 18 24 3651 1511 2899 71 115 201 202 199 1639 2595 210 189 1355 268 241 137 728 1899 193 192 325
139 143 181 261 342 349 373 433 622 623 686 700 745 812 868 951 987 990 1001 1016 1021 1046 1047 1069
888.jpg的第19项功能与709.jpg **的第139页相匹配
我使用openCV的sift算法，因此我为每个图像得到一个矩阵，其中包含 #Features x 128 ，我将它们保存为每个图像的.sift。
之后我使用openCV中的匹配算法并获得一个包含类似行的文件：
Image__2018-04-24__15-06-04.jpg Image__2018-04-24__15-06-11.jpg 955
22,27,33,49,59,65,72,92,97,100,101,105,106,112,116,120,123,126,127,129,137,141,142,143 ，144,147,152,159,160,169,179,186,188,207,223,233,250,261,265,266,268,269,292,296,297,299,300,302,304 ，306,308,309,312,315,318,326,327,328,329,331,333,334,337,342,348,350,358,361,364,365,369,376,380,382 ，383,385,386,396,403,404,407,408,410,411,417,424,427,428,429,431,432,434,435,437,438,441,443,445,446 ，449,450,451,452,455,458,464,465,467,468,470,471,474,477,478,480,481,482,486,488,489,490,493,494,495 ，496,499,501,504,505,506,509,510,511,512,514,515,516,524,526,527,528,529,530,531,533,536,538,539,541 ，547,549,550,554,556,560,561,563,565,567,573,574,575,577,581,582,585,587,589,590,591,596,599,602,605 ，610,615,616,617,618,629,630,631,633,659,665,670,671,678,680,681,682,683,686,690,692,694,698,703,706 ，709,710,7 21,723,735,736,737,739,740,742,743,744,745,746,747,748,750,751,752,753,754,756,757,759,760,761,762， 763,765,767,768,769,770,771,772,773,774,775,777,787,788,789,791,792,793,794,796,800,801,803,804,805， 806,809,810,811,812,814,815,817,818,819,820,821,822,825,828,829,831,834,835,836,838,839,840,841,843， 844,846,847,848,851,853,854,857,858,861,863,864,865,866,867,868,869,870,871,872,874,875,876,877,878， 879,882,883,885,886,888,889,890,894,895,897,898,901,902,905,907,909,910,912,913,914,916,917,918,919， 920,921,922,923,924,925,926,927,930,931,932,933,934,935,936,937,938,941,942,943,947,948,949,950,953， 954,957,958,959,960,961,964,965,968,969,970,971,972,973,974,976,977,978,979,980,981,982,983,985,987， 988,989,990,991,992,994,995,997,1001,1002,1003,1004,1005,1007,1008,1009,1010,1012,1013,1014,1015,1018,10 19,1020,1021,1022,1023,1025,1026,1027,1028,1029,1031,1033,1034,1038,1040,1041,1042,1044,1046,1047,1048,1050,1053,1055,1056， 1058,1060,1062,1065,1066,1067,1068,1070,1072,1074,1075,1076,1077,1078,1079,1081,1082,1083,1086,1087,1088,1089,1090,1092,1094， 1095,1096,1097,1099,1100,1102,1103,1105,1106,1107,1108,1109,1110,1111,1112,1113,1114,1115,1116,1117,1118,1122,1123,1124,1125， 1126,1133,1134,1135,1136,1137,1139,1141,1142,1143,1144,1145,1159,1161,1165,1167,1168,1169,1170,1173,1175,1176,1178,1179,1180， 1183,1185,1186,1187,1189,1190,1192,1197,1199,1200,1202,1203,1205,1206,1207,1209,1210,1211,1214,1215,1217,1218,1222,1223,1224， 1225,1227,1228,1229,1232,1234,1235,1238,1239,1240,1242,1243,1244,1249,1250,1251,1252,1253,1257,1261,1262,1263,1266,1268,1269， 1270,1271,1272,1273,1274,1275,1277,1278,1280,1281,1283,1284,1285,1286,1288,1289,1291， 1292,1293,1294,1295,1296,1298,1300,1301,1302,1303,1304,1307,1308,1309,1311,1312,1314,1315,1317,1318,1319,1320,1321,1323,1327， 1328,1329,1331,1332,1334,1335,1336,1337,1338,1339,1341,1343,1344,1345,1346,1347,1349,1350,1351,1352,1353,1354,1355,1356,1360， 1361,1362,1363,1366,1368,1372,1373,1374,1375,1376,1377,1378,1379,1380,1381,1382,1383,1384,1385,1386,1387,1388,1389,1390,1391， 1392,1393,1394,1396,1399,1401,1402,1403,1404,1406,1407,1408,1409,1411,1412,1414,1415,1416,1417,1418,1420,1421,1423,1425,1426， 1428,1429,1430,1431,1433,1436,1437,1438,1441,1442,1443,1444,1445,1446,1447,1449,1450,1452,1453,1454,1456,1457,1460,1462,1464， 1465,1467,1470,1471,1472,1473,1474,1475,1478,1480,1481,1483,1484,1485,1488,1490,1492,1493,1494,1495,1496,1497,1498,1499,1500， 1501,1502,1503,1504,1506,1508,1509,1510,1513,1514,1515,1516,1517,1518,1519,1520,1521 ，1523,1524,1525,1526,1527,1528,1531,1532,1534,1537,1539,1540,1541,1542,1543,1545,1547,1548,1549,1551,1552,1553,1554,1555,1557 ，1558,1560,1561,1563,1565,1566,1567,1568,1569,1570,1572,1573,1575,1576,1577,1578,1579,1581,1582,1583,1584,1585,1586,1587,1588 ，1590,1591,1593,1594,1597,1598,1599,1600,1601,1603,1604,1605,1608,1609,1610,1611,1612,1613,1614,1616,1618,1619,1620,1622,16223 ，1625,1627,1632,1633,1634,1635,1636,1637,1638,1649,1641,1643,1644,1648,1649,1650,1652,1653,1655,1656,1657,1658,1660,1661,1662 ，1663,1664,1665,1666,1667,1669,1670,1671,1672,1673,1674,1675,1677,1678,1679,1682,1683,1684,1685,1686,1689,1690,1692,1696,1697 ，1699,1700,1702,1705,1708,1710,1711,1712,1713,1714,1715,1717,1718,1719,1720,1723,1726,1727,1730,1731,1732,1734,1735,1736,1737 ，1739,1740,1741,1742,1744,1745,1747,1748,1749,1750,1752,1753,1754,1755,1756,1759,17 60,1761,1762,1763,1764,1765,1766,1767,1768,1769,1770,1771,1773,1775,1776,1778,1780,1782,1784,1785,1787,1788,1789,1790,1791， 1792,1793,1794,1795,1796,1797,1798,1799,1801,1802,1803,1806,1807,1811,1814,1815,1816,1819,1820,1821,1822,1823,1824,1825,1826， 1827年，1828年，1830年，1831年，1834年，1835年，1836年
所以我使用了sfm -import函数并插入了matches.txt和带有images和.sift文件的文件夹。但是我只得到一个空的n视图匹配文件（.nvm）
这个管道有什么问题吗？并有任何人暗示我如何工作？在我看来，我的格式有问题，但我看不出可能出错的地方
感谢您的帮助
最佳
最高