Chevron Left
Zurück zu Robotics: Perception

Bewertung und Feedback des Lernenden für Robotics: Perception von University of Pennsylvania

610 Bewertungen
169 Bewertungen

Über den Kurs

How can robots perceive the world and their own movements so that they accomplish navigation and manipulation tasks? In this module, we will study how images and videos acquired by cameras mounted on robots are transformed into representations like features and optical flow. Such 2D representations allow us then to extract 3D information about where the camera is and in which direction the robot moves. You will come to understand how grasping objects is facilitated by the computation of 3D posing of objects and navigation can be accomplished by visual odometry and landmark-based localization....


31. Jan. 2021

This course was truly amazing. It was challenging and I learned a lot of cool stuff. It would have been better if more animations were included in explaining complex concepts and equations.

31. März 2018

Outstanding Course! I could always count on Prof.Jianbo to crunch some of the most complex and confusing parts of the course into a much easier understandable language.

Filtern nach:

76 - 100 von 167 Bewertungen für Robotics: Perception

von jiqirenzhifu

27. Aug. 2017

I have gained a lot 。thank you

von Krishna C

17. Jan. 2022

Challenging and Fulfilling

von Qiu Q

19. Juni 2016

very good assignment!

von Terry Z

12. Jan. 2018

Very useful course!

von 徐恩科

5. Sep. 2018

very good course.


12. Sep. 2020

excellent course

von Ng B K

6. Aug. 2016

Love the course.

von Daniel W

2. Jan. 2017

Great course!

von Aryan A

20. Aug. 2018

Great course

von 张浩悦

31. Aug. 2017



von Xu D

27. Mai 2019

Love this!

von Meshal A

30. Mai 2017


von Bálint - H F

20. März 2019

Great !

von 杨镑镑

18. Aug. 2016


von Eduardo K d S

16. Sep. 2016

This course is great! There is a lot of information available, a wide range of topics are covered, some complex subjects are explained quite well and especially the Chinese professor makes it easy to understand them.

Still, there is room for improvement. Considering this is a 4-week course and the coverage of the material, sometimes it feels too squeezed and cramped together. This could be improved by providing access to more references to other materials to complement the studies. A bibliography for instance would be much welcome.

Also there are some annoying typos in the slides in the formulas and its derivations that can cost you some precious time to figure out, especially during the Matlab assignments.

These are the only reasons I don't give this course 5 stars, but it's definitely worthwhile. You will not regret it!

von Sourav G

20. Feb. 2018

It was really an interesting course and is recommended for those interested in Vision-based applications for their robots, especially dealing with motion estimation, visual odometry, visual SLAM, image matching using local point features (SIFT) etc. The course did help a lot in brushing up some concepts from undergrad and using them to create some amazing codes through assignments.

There are few things that can be improved, for example, some of the videos in the course lack proper explanation and it took a while to understand. Some of the quizzes comprise questions to which answers cannot be derived using the course content (AFAIU). The inverse depth parameterization-based direct pose estimation is not covered (e.g. as in LSD-SLAM).

von Liang M

17. Apr. 2017

A very good course in general. The materials and assignments are practical and the explanation of the instructors are clear. You are expect to gain a general knowledge about computer vision, camera calibration, and the usage of linear algebra in computer vision.

One thing that could be improved is that there is a big jump from week 2 to week 3 and also from week 3 to week 4. It's like a sophomore course at week 1 and week 2 and suddenly it jumps to a senior course in week 3 and a graduate course in week 4. It might be better to provide some supplementary materials in between.

von Berke

11. Okt. 2020

This course offers great way to begin with vision based applications such as visual odometry - slam i came this course because i wanted to learn math background about multiview geometry and algorithms such visual odometry and i learned a lot during the course (also made me realize that i need to study more math). The one setback about this course is that some part of this(week 2-3 mostly) course can be improved.

von Fernando C

17. Mai 2016

This course has a lot of interesting material regarding perception using cameras. The lectures focus on a wide range of topics, from the basics until camera pose estimation, epipolar geometry, optical flow and 3D motion. The explanations are very clear, and the Jacobian explanation using colors is excellent.

Negative points: sometimes the lectures are long, and the concepts are a little bit mixed.

von Qirui Z

6. Juni 2019

Still some errors in the homework PDF (The codes are all working though). And the curriculum seems a little bit redundant. Also hope there will be more emphasis on emerging applications like visual odometry and SLAM, in stead of spending too much time on the ancient geometry (It's not expected to always have a checkerboard in your image? :-) ).

von Matthew P

3. Mai 2018

The course was very detailed but perhaps a little too densely packed for a 4 week course. The instructors covered a very wide breadth of material in a very short time. I believe the instruction for the final assignment could be improved, but overall a good class introducing many important concepts in robotics and vision systems.

von Shaun K

4. Apr. 2017

Loved the lecture and materials. However, the course need more ACTIVE teaching staff and mentors. I had several questions regarding the materials but could not get any help from start to the end. It was the only specialization course that I had to move on without complete understanding of the materials.

von Lucila P

4. Nov. 2016

The course has a good structure. It covers interesting themas. The assignments are easy to understand. It takes more effort than 3 or 5 hours/week, the nomenclatur could be improved to be consistent. A couple of more examples would improve learning. The four week was hard ;)


8. Mai 2019

The course content is exceptionally well and material is well designed. Since most of the concepts are in-depth more support will be required in discussion forums. I feel the response from mentors is slow and have to wait days for their reply to clarify the doubts.

von Max B

9. Mai 2021

Wish there were more reading materials even it there weren't enough time to cover them. Too often, professors jumped from one concept to another without bridging.

Also, please fix typos in the videos and lecture notes.