Angeboten in den Modulen:
INF-AQUA, INF-BAS2, INF-BAS7, INF-D-940, INF-VERT2, INF-VERT7, IST-05-HS, MINF-04-HSWinterersemester 2015/16
Termin
Montag, 4.DS, 13:00 - 14:30 Uhr, ABP E010 (first time, other dates may be INF-2026 (Labor 2. Etage))Beginn: Mo, 12. Oktober 2015
Leitung
Prof. Carsten Rother, Konsultationen: siehe ThemaBeschreibung
This seminar is focused on current Computer Vision topics and literature that are in the interest of our group.In everey seminar there is a presenter and a reviewer of a paper. You have to be both once. The timeline then is:
- problem presentation (5 min.)
- discussion (10min.)
- presentation of the solution of the paper (15min.)
- discussion (5min.)
- review (10 min.)
- we will have two papers per seminar
- you have to hand in your slides and your review
- preparing and giving a scientific presentation, hints: 1, 2
- leading a discussion
- understanding current scientific literature
- learn to gain a critical but fair look at papers by writing and presenting a review, see also ICCV reviewer instructions
All talks are in English.
Einschreibung:
Einschreibung über jExam, Themenwahl über TUD-email an Holger.HeidrichTerminvergabe zum ersten Seminar entsprechend der Einschreibreihenfolge.
Inhalt: matching and scene understanding
Themenangebot:- Weakly Supervised Semantic Segmentation for Social Images, Wei Zhang, Sheng Zeng, Dequan Wang, Xiangyang Xue, CVPR 2015 ( presenter: Omid , reviewer: ) (mentor Omid )
- Edge Boxes: Locating Object Proposals from Edges, cancelled (Amir leaves us unfortunately)
- Human-level control through deep reinforcement learning, ( presenter: Alex Krull , reviewer: ) (mentor Alex Krull)
- Superdifferential Cuts for Binary Energies, Taniai, Tatsunori, Yasuyuki Matsushita, and Takeshi Naemura, CVPR 2015. ( presenter: reviewer: ) (mentor Bogdan Savchynskyy)
- Depth from Focus with Your Mobile Phone, Supasorn Suwajanakorn, Carlos Hernandez and Steven M. Seitz, CVPR 2015 ( presenter: reviewer: ) (mentor Anita Sellent)
- FlowWeb: Joint Image Set Alignment by Weaving Consistent, Pixel-wise Correspondences, ( presenter: Masa, reviewer: Weihao ) (mentor Masa)
- Incremental Dense Semantic Stereo Fusion for Large-Scale Semantic Scene Reconstruction, ICRA2015 ( presenter: reviewer: ) (mentor Michael Ying Yang)
- Can Humans Fly? Action Understanding with Multiple Classes of Actors., C. Xu, S.-H. Hsieh, C. Xiong, and J. J. Corso. CVPR 2015 ( presenter: Weihao Li , reviewer: Siva ) (mentor Weihao Li)
- Describing the Scene as a Whole: Joint Object Detection, Scene Classification and Semantic Segmentation, J. Yao, S. Fidler and R. Urtasun, 2012 (Carsten Rother)
- A Local Algorithm for the Computation of Image Velocity via Constructive Interference of Global Fourier Components,Babette Dellen, Florentin Wörgötter, International Journal of Computer Vision, 2011 (Holger Heidrich)
- Tractable Algorithms for Robust Model Estimation, Enqvist, Olof and Ask, Erik and Kahl, Fredrik and Åström, Kalle, (2014), (Holger Heidrich)
- Shortest Paths with Higher-Order Regularization, Ulén, Johannes and Strandmark, Petter and Kahl, Fredrik, (2015), ( presenter: Christian Weigert , reviewer: ) (Holger Heidrich)
- Extremely Low Bit-Rate Nearest Neighbor Search Using a Set Compression Tree, R. Arandjelović, A. Zisserman , (2013), ( presenter: Maximilian Knespel , reviewer: ) (Holger Heidrich)
- more may come
THEMA | TERMIN | VORTRAGENDER |
---|---|---|
Einführung, FlowWeb: Joint Image Set Alignment by Weaving Consistent, Pixel-wise Correspondences | 12.10.2015 | Holger Heidrich presenter: Masa, reviewer: Weihao |
Extremely Low Bit-Rate Nearest Neighbor Search Using a Set Compression Tree | 09.11.2015 | Maximilian Knespel, review: Christian Weigert |
Can Humans Fly? Action Understanding with Multiple Classes of Actors. | 09.11.2015 | Weihao, review: Siva |
Describing the Scene as a Whole: Joint Object Detection, Scene Classification and Semantic Segmentation | 16.11.2015 | Siva, review: Omid |
Weakly Supervised Semantic Segmentation for Social Images | 23.11.2015 | Omid, review: Hassan |
Shortest Paths with Higher-Order Regularization | 23.11.2015 | Christian Weigert, review: Maximilian Knespel |
Human-level control through deep reinforcement learning | 30.11.2015 | Alex Krull, review: Alex Kirillov |
tba | 30.11.2015 | Alex Kirillov, review: Alex Krull, |
tba | 07.12.2015 | Hassan, review: |