1LOVESJohnny-CSDN博客

原创 CV | SCNN: Spatial CNN for Traffic Lane Detection 车道线检测算法SCNN“最全的”学习笔记

题外话：可能因为是节假日，所以这篇文章的标题非常的标题党哈哈哈哈哈~ 回归严肃，过去一段时间花了比较多的精力学习这个算法，所以想借此空闲时间，好好总结和复盘一下。感兴趣的朋友们，欢迎评论区讨论，如果有理解不对的地方也欢迎大家指正。前言车道线检测在自动驾驶感知任务中占据重要的一席，它可以辅助自动驾驶的车道保持，同时也可以为车辆定位等工作提供较强的先验信息。目前，常见的方法大抵...

2019-04-06 22:35:53 7673 39

原创 LeetCode｜501. Find Mode in Binary Search Tree

时隔若干年后，我又开始刷leetcode了，看看这次会刷多少。

2024-04-09 22:21:57 366 1

原创面试经验｜PlayStation全球交互娱乐组——技术经理面

面试经验第二贴，索尼旗下的PS英国组。自我介绍部分真是每个面试必做的事情，可以提前准备好，重点突出自己的经历和优势，以及适当引导面试官往比较匹配的项目经验上提问。

2024-04-07 00:40:49 775

原创面试经验｜创业公司CTO面和技术总监面

权重激活函数的输出（前向的结果）梯度这个的答案是最容易受精度影响的是权重，最好是用最高的数据精度表示权重。梯度的影响相对最小，因为每次求反向传播更新的权值会乘以学习率learning rate使得更新值更低。

2024-04-02 05:09:22 494

最惊喜的事情是，在平稳地交完满意的作业后，我终于迈出上瑜伽课的一步。在感受完身体的拉伸和放松后，身体释放出快乐的要素，我就遇到了阳光、朋友们，和意外的午饭。经常被身边这群朋友们感动和教育，尽管我常常说长他们几岁，但实际上他们都在用实际行动教会我如何去爱，如何支持，如何信任，如何追求和平衡目标及状态。我看见的是希望，她对自己的希望，同时心态放松，悠闲地期待着。为了徒步旅行做准备，一个人骑车去迪卡侬，在回家的路上突然一阵强烈的伤感。印象中，一月的某天醒来，我突然有一种深深的释然，也惊奇于之前自己的沮丧。

2024-03-01 07:58:19 184

原创 CUDA | Writing and Compiling a CUDA Code (中英版)

CUDA is the extension of C++.

2024-01-31 23:48:59 535

原创随笔 | 写在年末冬日的夜里

原来设想的十二月，交完两个大作业的我，也许会是宅在家里看一个星期的圣诞电影；然而，万万没想到，以为一切都在预期中的轻松且专注的十二月，发生了意料之外的戏剧情节。言简意赅，游离恍神，清醒过来，接下来的这条路我要一个人走了。我们的问题一直非常明显，我向往天空、大海、辽阔的可能性，他需要稳定、确定性、以及物质的要求。无论是以什么样的方式，结束了这段关系，时至今日，我已经可以完全理解。现在，即使有阵痛，我也不知道这个阵痛会持续多久，我也满心期待。我都无法想象，如果没有你们，我究竟会体验如何的伤痛。

2023-12-30 04:10:42 1214

原创随笔 | 写在第一个学期中后，赶ddl两眼发黑，希望是黎明前的黑暗

如果有人问，英国的硕士一年水不水。我只想说：水的话，来试试吧！一句话总结过去三周的感受：度日如年。但也好久，没有这么酣畅淋漓地学习了。

2023-11-17 08:30:20 115

原创随笔 | 写在剑桥学习一个月的这一天

而这一句，不是轻而易举的言语，它需要我一步一个脚印，在完成一个个作业，在一次次打开自己，在一次次抚慰自己，然后继续往前走的这些时刻，给我自己鼓励和肯定。哪知道课程导师比我想象中的还要豁达，我们只聊了2分钟，然后他又展现出他迷人的洒脱的大大的微笑，跟我说：完全可以理解！工作中，我以为自己有很多不能控制的、选择的部分，是不是也可以放轻松一些，尽情顺应自己的想法去选择吧。窗外的风景让我想起《苔丝》的画面，一次次感觉四月末看的这部电影，也在给我什么安慰和力量。尽管屋内面积不大，对我来说，正好足够了。

2023-10-26 04:49:10 198

原创随笔 | 写在八月末的这一天

八月，拉开了序章。尽量整理和梳理好当前的工作，希望可以达成一个milestone。也因为正式提出要离开，收获了很多认可的同事的认可。超级感动。虽然反复提醒自己不要在意他人的评价，而信赖的合作过的同事的评价还是会给我很大的鼓励。记在这篇文章里，只希望日后怀疑自己的时候，可以打起精神来。虽然提出离开非常艰难，但还是很感动老板的关心和支持。同时，非常荣幸和这样的团队和公司，共同经历一次次的技术浪潮。好好告别，继续向前。

2023-09-01 00:04:04 108

原创随笔 | 写在七月末的这一天

这篇本应该是七月底写完，各种原因，在八月末才有时间完成。还是希望可以多记录呀~

2023-08-31 22:39:10 235

原创随笔 | 写在六月末的这一天

你好，七月。在我最喜欢的月份，祝愿他平安、顺利。

2023-06-30 22:35:50 219

原创随笔 | 写在五月末的这一天

又要用那个经典的开头：转眼，五月一瞬间就过去了。题外话：昨天一时兴起看了一部有点吓人的美剧，导致我昨天今天以来有点恍惚。看来还是不能因为一时冲动尝试自己不擅长的事情，尤其是看剧。题外话结束，进入月度碎碎念正题。

2023-05-31 23:59:33 433

原创随笔 | 如何准备奖学金面试

打招呼、介绍现状介绍highlight背景：2-3句话结束语。

2023-04-30 16:47:08 676 1

原创随笔 | 写在3月的最后一天

说来也巧，明天我又要启程去SH了。距离去年的长时间居家，已经过去了整整一年。晚上听着近期最爱的来都来了，正好听到去年同时期的一期节目。主播们一起哭，我的眼泪也不禁流下来。原来情绪都是正常的，因为我们的情绪弹簧功能正常。哭也不是软弱的表现。哭完，我们就来迎接新的一个月吧。另外，明天是我回归入职这家公司4周年。这四年，弹指一挥间。很幸运见证了它的一次次震撼。继续加油呀~他刚才跟我说，SH的樱花开的很美。再见，三月；你好，四月。

2023-03-31 22:56:09 378

原创 Package | 解决 graphviz.backend.execute.ExecutableNotFound: failed to execute PosixPath(‘dot‘)

graphviz.backend.execute.ExecutableNotFound: failed to execute PosixPath('dot'), make sure the Graphviz executables are on your systems' PATH

2023-01-31 23:15:48 2938 3

原创 Package | 解决 RuntimeError: Numpy is not available. Failed to initialize NumPy against version 0xf

UserWarning: Failed to initialize NumPy: module compiled against API version 0xf but this version of numpy is 0xd (Triggered internally at /opt/pytorch/pytorch/torch/csrc/utils/tensor_numpy.cpp:68.)

2022-11-02 16:16:26 6243 2

原创随笔 | 写在十一月的第一天

整个十月没有如愿输出技术文章，在十一月的第一天，简单回顾一下过去的十月。

2022-11-01 23:55:55 212

原创随笔 | 写在九月末的这一天

这个月又没来得及更新技术博客，时间走的越来越快了呢（bushi（是的，对于第三个关键词完全没有自责的意思。（（bushi。

2022-09-30 23:57:09 482

原创 Package | 解决Could NOT find GLEW (missing: GLEW_INCLUDE_DIRS GLEW_LIBRARIES)

实验环境：Ubuntu 18.04+Docker container，出现GLEW报错信息如Could NOT find GLEW (missing: GLEW_INCLUDE_DIRS GLEW_LIBRARIES)。

2022-08-31 14:25:46 3998

原创 Package | 解决 module ‘cv2‘ has no attribute ‘gapi_wip_gst_GStreamerPipeline‘

由于这个问题出现了两回，决定记录一下。实验背景是使用opencv python库进行数据预处理，遇到报错信息如下：“import cv2”

2022-08-22 15:19:39 35177 10

原创随笔 | 写在七月末的这一天

这个月过的太快，快到都来不及写技术博客（bushi。在月末快要过去之前，还是照常打开了博客。想着唠叨几句，试图记录些什么。

2022-07-31 23:49:44 175

原创 LeetCode | 876. Middle of the Linked List

Given the head of a singly linked list, return the middle node of the linked list.If there are two middle nodes, return the second middle node.Input: head = [1,2,3,4,5]Output: [3,4,5]Explanation: The middle node of the list is node 3.Input: head = [1,2,3

2022-06-30 23:54:46 177

原创 LeetCode | 784. Letter Case Permutation

Given a string s, you can transform every letter individually to be lowercase or uppercase to create another string.Return a list of all possible strings we could create. Return the output in any order.Input: s = “a1b2”Output: [“a1b2”,“a1B2”,“A1b2”,“A1B2”

2022-06-30 23:42:45 130

原创 LeetCode | 771. Jewels and Stones

You’re given strings jewels representing the types of stones that are jewels, and stones representing the stones you have. Each character in stones is a type of stone you have. You want to know how many of the stones you have are also jewels.Letters are ca

2022-06-30 23:40:21 147

原创 LeetCode | 725. Split Linked List in Parts

Given the head of a singly linked list and an integer k, split the linked list into k consecutive linked list parts.The length of each part should be as equal as possible: no two parts should have a size differing by more than one. This may lead to some pa

2022-06-30 23:37:06 162

原创 LeetCode | 719. Find K-th Smallest Pair Distance

The distance of a pair of integers a and b is defined as the absolute difference between a and b.Given an integer array nums and an integer k, return the kth smallest distance among all the pairs nums[i] and nums[j] where 0

2022-06-30 23:34:00 1391

原创 LeetCode | 707. Design Linked List

Design your implementation of the linked list. You can choose to use a singly or doubly linked list.A node in a singly linked list should have two attributes: val and next. val is the value of the current node, and next is a pointer/reference to the next

2022-06-30 23:30:17 117

原创 LeetCode | 693. Binary Number with Alternating Bits

Given a positive integer, check whether it has alternating bits: namely, if two adjacent bits will always have different values.

2022-05-31 21:27:13 105

原创 LeetCode | 680. Valid Palindrome II

Given a string s, return true if the s can be palindrome after deleting at most one character from it.

2022-05-31 21:24:44 131

原创 LeetCode | 665. Non-decreasing Array

Given an array nums with n integers, your task is to check if it could become non-decreasing by modifying at most one element.We define an array is non-decreasing if nums[i]

2022-05-31 21:21:23 148

原创 LeetCode | 658. Find K Closest Elements

Given a sorted integer array arr, two integers k and x, return the k closest integers to x in the array. The result should also be sorted in ascending order.An integer a is closer to x than an integer b

2022-05-31 21:18:54 102

原创 LeetCode | 647. Palindromic Substrings

Given a string s, return the number of palindromic substrings in it.A string is a palindrome when it reads the same backward as forward.A substring is a contiguous sequence of characters within the string.

2022-05-31 21:15:53 110

原创 LeetCode | 639. Decode Ways II

.题目A message containing letters from A-Z can be encoded into numbers using the following mapping:‘A’ -> “1”‘B’ -> “2”…‘Z’ -> “26”To decode an encoded message, all the digits must be grouped then mapped back into letters using the reverse of

2022-04-30 22:07:14 321

原创 LeetCode | 637. Average of Levels in Binary Tree

.题目Given the root of a binary tree, return the average value of the nodes on each level in the form of an array. Answers within 10-5 of the actual answer will be accepted.Example 1:Input: root = [3,9,20,null,null,15,7]Output: [3.00000,14.50000,11.000

2022-04-30 22:02:43 192

原创 LeetCode | 628. Maximum Product of Three Numbers

.题目Given an integer array nums, find three numbers whose product is maximum and return the maximum product.Example 1:Input: nums = [1,2,3]Output: 6Example 2:Input: nums = [1,2,3,4]Output: 24Example 3:Input: nums = [-1,-2,-3]Output: -6Constraint

2022-04-30 21:57:43 154

原创随笔 | 武汉大学管理课笔记第一课

前言如何打造自己的品牌？比如，如何将武汉大学的管理课程做好，或者如何为在武汉大学中学习的同学打造品牌？– 首先，好好学习理论，才能从实践中掌握和理解。利用好学校的人文科学优势。– 奋斗。– 良好的为人处事的原则。...

2022-04-16 16:41:51 450

原创 LeetCode | 617. Merge Two Binary Trees

.题目You are given two binary trees root1 and root2.Imagine that when you put one of them to cover the other, some nodes of the two trees are overlapped while the others are not. You need to merge the two trees into a new binary tree. The merge rule is th

2022-03-31 22:03:08 212

原创 LeetCode | 605. Can Place Flowers

.题目You have a long flowerbed in which some of the plots are planted, and some are not. However, flowers cannot be planted in adjacent plots.Given an integer array flowerbed containing 0’s and 1’s, where 0 means empty and 1 means not empty, and an intege

2022-03-31 21:58:18 359

原创 LeetCode | 872. Leaf-Similar Trees

.题目Consider all the leaves of a binary tree, from left to right order, the values of those leaves form a leaf value sequence.For example, in the given tree above, the leaf value sequence is (6, 7, 4, 9, 8).Two binary trees are considered leaf-similar

2022-03-31 21:46:01 1139

Semantic Co-segmentation in Videos

Discovering and segmenting objects in videos is a challenging task due to large variations of objects in appearances, deformed shapes and cluttered backgrounds. In this paper, we propose to segment objects and understand their visual semantics from a collection of videos that link to each other, which we refer to as semantic co-segmentation. Without any prior knowledge on videos, we rst extract semantic objects and utilize a tracking-based approach to generate multiple object-like tracklets across the video. Each tracklet maintains temporally connected segments and is associated with a predicted category. To exploit rich information from other videos, we collect tracklets that are assigned to the same category from all videos, and co-select tracklets that belong to true objects by solving a submodular function. This function accounts for object properties such as appearances, shapes and motions, and hence facilitates the co-segmentation process. Experiments on three video object segmentation datasets show that the proposed algorithm performs favorably against the other state-of-the-art methods.

2017-01-05

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials

Most state-of-the-art techniques for multi-class image segmentation and labeling use conditional random fields defined over pixels or image regions. While regionlevel models often feature dense pairwise connectivity, pixel-level models are considerably larger and have only permitted sparse graph structures. In this paper, we consider fully connected CRF models defined on the complete set of pixels in an image. The resulting graphs have billions of edges, making traditional inference algorithms impractical. Our main contribution is a highly efficient approximate inference algorithm for fully connected CRF models in which the pairwise edge potentials are defined by a linear combination of Gaussian kernels. Our experiments demonstrate that dense connectivity at the pixel level substantially improves segmentation and labeling accuracy.

2017-01-02

Feature Space Optimization for Semantic Video Segmentation

We present an approach to long-range spatio-temporal regularization in semantic video segmentation. Temporal regularization in video is challenging because both the camera and the scene may be in motion. Thus Euclidean distance in the space-time volume is not a good proxy for correspondence. We optimize the mapping of pixels to a Euclidean feature space so as to minimize distances between corresponding points. Structured prediction is performed by a dense CRF that operates on the optimized features. Experimental results demonstrate that the presented approach increases the accuracy and temporal consistency of semantic video segmentation.

2017-01-02

An Experimental Comparison of Min-Cut/Max-Flow Algorithms

最大流/最小割算法的简介，理解常用最大流最小割概念的文献，值得学习。 minimum cut/maximum flow algorithms on graphs emerged as an increasingly useful tool for exact or approximate energy minimization in low-level vision. The combinatorial optimization literature provides many min-cut/max-flow algorithms with different polynomial time complexity. Their practical efficiency, however, has to date been studied mainly outside the scope of computer vision. The goal of this paper is to provide an experimental comparison of the efficiency of min-cut/max flow algorithms for applications in vision. We compare the running times of several standard algorithms, as well as a new algorithm that we have recently developed. The algorithms we study include both Goldberg-Tarjan style “push-relabel” methods and algorithms based on Ford- Fulkerson style “augmenting paths.” We benchmark these algorithms on a number of typical graphs in the contexts of image restoration, stereo, and segmentation. In many cases, our new algorithm works several times faster than any of the other methods, making near real-time performance possible. An implementation of our max-flow/min-cut algorithm is available upon request for research purposes.

2016-08-24

Object Recognition with Color Cooccurrence Histograms

We use the color cooccurrence histogram (CH) for recognizing objects in images. The color CH keeps track of the number of pairs of certain colored pixels that occur at certain separation distances in image space. The color CH adds geometric information to the normal color histogram, which abstracts away all geometry. We compute model CHs based on images of known objects taken from different points of view. These model CHs are then matched to subregions in test images to find the object. By adjusting the number of colors and the number of distances used in the CH, we can adjust the tolerance of the algorithm to changes in lighting, viewpoint, and the flexibility of the object. We develop a mathematical model of the algorithm’s false alarm probability and use this as a principled way of picking most of the algorithm’s adjustable parameters. We demonstrate our algorithm on different objects, showing that it recognize objects in spite of confusing background clutter, partial occlusions, and flexing of the object.

2016-07-21

(matlab) Seam_Carving

Shai Avidan Mitsubishi Electric Research Labs, SeamCarving for Content-Aware Image Resizing, 2007 源码资料；图像智能缩放的经典算法；值得学习。

2016-07-15

Robust Image Retargeting via Axis

图像智能缩放（Image Retargeting）方向的优秀文章！值得学习！

2016-07-14

Seam Carving for Content-Aware Image Resizing

传统的等比例缩放技术，直接将图像整体缩放为目标尺寸，这样做通常会破坏图像内容的比例关系，因为会出现目标长宽比与源图像长宽比不一的情况，图像内容因此会受到损失，另一方面如果目标尺寸较小而源图像的重要信息区域占整幅图像比例不大的情况下进行等比例缩放的结果是，图像信息同样会损失严重。图像智能缩放算法不同于传统的缩放技术，该算法根据图像内容的分析得出图像重要信息的分布情况，然后尽可能保留重要信息的同时，对图像进行连续且单调的像素路径的移除或插入操作，最后在保证图像矩形格式不变的前提下，缩放生成满足目标图像尺寸。

2016-07-14

达拉然的守护

任务为了帮助萧邦顺利通过跨省冒险，助华尔兹反败为胜，大家要去捡尽可能多的粮草，你可以上下左右四种方式行走，找到粮食便可以捡起，最后上报你能拿到的粮食的最大数量。由于阿尔塞斯的粮仓诡异莫测，你会被传送到粮仓的任意一个地方。输入阿尔塞斯有多个粮仓，所以输入数据有多组，每一组数据先输入两个数字，分别代表粮仓的长x和宽y（没有greedisgood指令，所以资源有限，x和y<=2000）然后输入相应大小的粮仓构造，由y行x列组成。当x和y都为0的时候输入结束。其中：“.”代表是粮食，“#”代表是防护网（意味着你不可以跨越），“@”代表你被传送的位置（传送到的位置上面有粮食）公开的第一个用例即是格式。可以自行理解。输出在你被传送到的位置上，可以捡到的最大粮食数量。

2013-04-02

堆排序代码

实验要求：用堆排序算法按关键字递减的顺序排序。程序输入：待排序记录数（整数）和待排序记录（整数序列）；程序输出：建堆结果和建堆后第一、第二次筛选结果。（注：待排序记录数大于等于3）

2013-01-13

空空如也

TA创建的收藏夹 TA关注的收藏夹

TA关注的人