[course site]
Xavier Giro-i-Nieto
xavier.giro@upc.edu
Associate Professor
Universitat Politecnica de Catalunya
Technical University of Catalonia
Video Analysis
Day 2 Lecture 2
#DLUPC
Acknowledgments
2
Víctor Campos Alberto Montes
Outline
1. Recognition
2. Optical Flow
3. Object Tracking
3
Recognition
Demo: Clarifai
MIT Technology Review : “A start-up’s Neural Network Can Understand Video” (3/2/2015)
4
Figure: Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. (2014, June). Large-scale video classification with
convolutional neural networks. In Computer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on (pp. 1725-1732). IEEE.
5
Recognition
6
(Slides by Dídac Surís) Abu-El-Haija, Sami, Nisarg Kothari, Joonseok Lee, Paul Natsev, George Toderici, Balakrishnan Varadarajan, and Sudheendra
Vijayanarasimhan. "Youtube-8m: A large-scale video classification benchmark." arXiv preprint arXiv:1609.08675 (2016). [project]
Activity Recognition: Datasets
7
Recognition
Figure: Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D
convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
8
Recognition
Figure: Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning
spatiotemporal features with 3D convolutional networks." ICCV 2015
Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L.
Large-scale video classification with convolutional neural networks. CVPR 2014
Slides extracted from ReadCV seminar by Victor Campos 9
Recognition: DeepVideo
10
Recognition: DeepVideo: Demo
Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video
classification with convolutional neural networks. CVPR 2014
11
Recognition: DeepVideo: Architectures
Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video
classification with convolutional neural networks. CVPR 2014
12
Recognition: DeepVideo: Features
Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video
classification with convolutional neural networks. CVPR 2014
13
Recognition: DeepVideo: Multiscale
Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video classification with convolutional neural networks.
CVPR 2014
14
Recognition: DeepVideo: Results
Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video classification with convolutional neural networks.
CVPR 2014
15
Jeffrey Donahue, Lisa Anne Hendricks, Sergio Guadarrama, Marcus Rohrbach, Subhashini Venugopalan, Kate Saenko,
Trevor Darrel. Long-term Recurrent Convolutional Networks for Visual Recognition and Description, CVPR 2015. code
Activity Recognition: Frames + LSTM
16
Yue-Hei Ng, Joe, Matthew Hausknecht, Sudheendra Vijayanarasimhan, Oriol Vinyals, Rajat Monga, and
George Toderici. "Beyond short snippets: Deep networks for video classification." CVPR 2015
Activity Recognition: Frames + Optical Flow + LSTM
17
Recognition
Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning
spatiotemporal features with 3D convolutional networks." ICCV 2015
18
Simonyan, Karen, and Andrew Zisserman. "Two-stream convolutional networks for action recognition in
videos." NIPS 2014.
Recognition: Two stream
Two CNNs in paralel:
● One for RGB images
● One for Optical flow (hand-crafted features)
Fusion after the softmax layer
19Feichtenhofer, Christoph, Axel Pinz, and Andrew Zisserman. "Convolutional two-stream network fusion for video action recognition." CVPR 2016. [code]
Recognition: Two stream
Two CNNs in paralel:
● One for RGB images
● One for Optical flow (hand-crafted features)
Fusion at a convolutional layer
20
Recognition
Figure: Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D
convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
21
Recognition: C3D
Figure: Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning
spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International
Conference on Computer Vision, pp. 4489-4497. 2015
22
Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks."
In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
Recognition: C3D: Demo
23
Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks."
In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
Recognition: C3D: Temporal dimension
3D ConvNets are more suitable for spatiotemporal feature learning compared to 2D ConvNets
Temporal depth
2D ConvNets
24
Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks."
In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
A homogeneous architecture with small 3 × 3 × 3 convolution kernels in all layers is among the best
performing architectures for 3D ConvNets
Recognition: C3D: Temporal dimension
25
Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks."
In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
Recognition: C3D: Architecture
Feature
vector
26
Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks."
In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
Recognition: C3D: Feature vector
16-frame clip
16-frame clip
16-frame clip
16-frame clip
...
Average
4096-dimvideodescriptor
4096-dimvideodescriptor
L2 norm
BSc
thesis
http://activity-net.org/ 27
Temporal Activity Detection
BSc
thesis
Videos
Activity Classification
Longboarding
28
Temporal Activity Detection
BSc
thesis
Videos
Activity Temporal Localization
Longboarding
29
Temporal Activity Detection
30
Shou, Zheng, Dongang Wang, and Shih-Fu Chang. "Temporal action localization in untrimmed videos via multi-stage cnns." CVPR 2016.
(Slidecast and Slides by Alberto Montes)
Recognition: Localization
31
Recognition: Localization
(Slidecast and Slides by Alberto Montes) Shou, Zheng, Dongang Wang, and Shih-Fu Chang. "Temporal action localization in
untrimmed videos via multi-stage cnns." CVPR 2016.
32
Recognition: Localization
(Slidecast and Slides by Alberto Montes) Shou, Zheng, Dongang Wang, and Shih-Fu Chang. "Temporal action localization in
untrimmed videos via multi-stage cnns." CVPR 2016.
BSc
thesis
Neural Network
Activity
33
Temporal Activity Detection
BSc
thesis
Activity
CNN RNN+
34
Temporal Activity Detection
Figure: Tran, Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning
spatiotemporal features with 3D convolutional networks." CVPR 2015
3D Convolutions over sets of 16 frames...
35
Temporal Activity Detection
BSc
thesis
36
Temporal Activity Detection
BSc
thesis
mAP = 0.5938 mAP = 0.5492 mAP = 0.5635
Deeper networks present overfitting
37
Temporal Activity Detection
BSc
thesis
38
Temporal Activity Detection
BSc
thesis
39
Temporal Activity Detection
BSc
thesis
40
Temporal Activity Detection
BSc
thesis
Ground Truth:
Playing water polo
Prediction:
0.765 Playing water polo
0.202 Swimming
0.007 Springboard diving
41
Temporal Activity Detection
BSc
thesis
Ground Truth:
Hopscotch
Prediction:
0.848 Running a marathon
0.023 Triple jump
0.022 Javelin throw
42
Temporal Activity Detection
BSc
thesis
43
Temporal Activity Detection
A. Montes, Salvador, A., Pascual-deLaPuente, S., and Giró-i-Nieto, X., “Temporal
Activity Detection in Untrimmed Videos with Recurrent Neural Networks”, in 1st
NIPS Workshop on Large Scale Computer Vision Systems 2016 (best poster award)
44
Temporal Activity Detection
A. Montes, Salvador, A., Pascual-deLaPuente, S., and Giró-i-Nieto, X., “Temporal
Activity Detection in Untrimmed Videos with Recurrent Neural Networks”, in 1st
NIPS Workshop on Large Scale Computer Vision Systems 2016 (best poster award)
Outline
1. Recognition
2. Optical Flow
3. Object Tracking
45
Optical Flow
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 46
Optical Flow: FlowNet
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. ICCV 2015 47
Optical Flow: FlowNet
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 48
End to end supervised learning of optical flow.
Optical Flow: FlowNet (contracting)
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 49
Option A: Stack both input images together and feed them through a generic
network.
Optical Flow: FlowNet (contracting)
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 50
Option B: Create two separate, yet identical processing streams for the two images
and combine them at a later stage.
Optical Flow: FlowNet (contracting)
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 51
Correlation layer:
Convolution of data patches from the layers to combine.
Option B: Create two separate, yet identical processing streams for the two images
and combine them at a later stage.
Optical Flow: FlowNet (expanding)
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 52
Upconvolutional layers: Unpooling features maps + convolution.
Upconvolutioned feature maps are concatenated with the corresponding map from the contractive part.
Optical Flow: FlowNet
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. ICCV 2015 53
Since existing ground truth datasets are not sufficiently large to train a Convnet, a synthetic Flying Dataset is
generated… and augmented (translation, rotation, scaling transformations; additive Gaussian noise; changes in
brightness, contrast, gamma and color).
Convnets trained on these unrealistic data generalize well to existing datasets such as Sintel and KITTI.
Data
augmentation
Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning
Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 54
Optical Flow: FlowNet
Outline
1. Recognition
2. Optical Flow
3. Object Tracking
55
Object tracking: Deep but not CNN
56
Wang, Naiyan, and Dit-Yan Yeung. "Learning a deep compact image representation for visual tracking."
NIPS 2013.
Offline learning: Robust and generic features are learning by training a stacked denoising autoencoder on
auxiliary images.
Online learning: Encoder part of the autoencode + classification neural network
Object tracking: MDNet
57
Nam, Hyeonseob, and Bohyung Han. "Learning multi-domain convolutional neural networks for visual tracking." ICCV VOT Workshop (2015)
Object tracking: MDNet: Architecture
58
Nam, Hyeonseob, and Bohyung Han. "Learning multi-domain convolutional neural networks for visual tracking." ICCV VOT Workshop (2015)
Domain-specific layers are used during training for each sequence, but are replaced by a single one at test
time.
Object tracking: MDNet: Online update
59
Nam, Hyeonseob, and Bohyung Han. "Learning multi-domain convolutional neural networks for visual tracking." ICCV VOT Workshop (2015)
MDNet is updated online at test
time with hard negative mining,
that is, selecting negative
samples with the highest positive
score.
Object tracking: FCNT
60
Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." ICCV 2015 [code]
Object tracking: FCNT
61
Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." CVPR 2015 [code]
Focus on conv4-3 and conv5-3 of VGG-16 network pre-trained for ImageNet image classification.
conv4-3 conv5-3
Object tracking: FCNT: Specialization
62
Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE
International Conference on Computer Vision, pp. 3119-3127. 2015 [code]
Most feature maps in VGG-16 conv4-3 and conv5-3 are not related to the foreground regions in a tracking
sequence.
Object tracking: FCNT: Localization
63
Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE
International Conference on Computer Vision, pp. 3119-3127. 2015 [code]
Although trained for image classification, feature maps in conv5-3 enable object localization…
...but is not discriminative enough to different objects of the same category.
Object tracking: Localization
64
Zhou, Bolei, Aditya Khosla, Agata Lapedriza, Aude Oliva, and Antonio Torralba. "Object detectors emerge in deep scene cnns." ICLR 2015.
Other works have shown how features maps in convolutional layers allow object localization.
Object tracking: FCNT: Localization
65
Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE
International Conference on Computer Vision, pp. 3119-3127. 2015 [code]
On the other hand, feature maps from conv4-3 are more sensitive to intra-class appearance variation…
conv4-3 conv5-3
Object tracking: FCNT: Architecture
66
Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE
International Conference on Computer Vision, pp. 3119-3127. 2015 [code]
SNet=Specific Network (online update)
GNet=General Network (fixed)
Object tracking: FCNT: Results
67
Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE
International Conference on Computer Vision, pp. 3119-3127. 2015 [code]
Object tracking: ROLO
68
Ning, Guanghan, Zhi Zhang, Chen Huang, Zhihai He, Xiaobo Ren, and Haohong Wang. "Spatially Supervised Recurrent Convolutional Neural
Networks for Visual Object Tracking." IEEE International Symposium on Circuits and Systems, 2017
Object tracking: ROLO
69
Ning, Guanghan, Zhi Zhang, Chen Huang, Zhihai He, Xiaobo Ren, and Haohong Wang. "Spatially Supervised Recurrent Convolutional Neural
Networks for Visual Object Tracking." arXiv preprint arXiv:1607.05781 (2016)
Object tracking: CFNet
70
Valmadre, Jack, Luca Bertinetto, João F. Henriques, Andrea Vedaldi, and Philip HS Torr. "End-to-end
representation learning for Correlation Filter based tracking." CVPR 2017
Questions?

Video Analysis (D4L2 2017 UPC Deep Learning for Computer Vision)

  • 1.
    [course site] Xavier Giro-i-Nieto xavier.giro@upc.edu AssociateProfessor Universitat Politecnica de Catalunya Technical University of Catalonia Video Analysis Day 2 Lecture 2 #DLUPC
  • 2.
  • 3.
    Outline 1. Recognition 2. OpticalFlow 3. Object Tracking 3
  • 4.
    Recognition Demo: Clarifai MIT TechnologyReview : “A start-up’s Neural Network Can Understand Video” (3/2/2015) 4
  • 5.
    Figure: Karpathy, A.,Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. (2014, June). Large-scale video classification with convolutional neural networks. In Computer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on (pp. 1725-1732). IEEE. 5 Recognition
  • 6.
    6 (Slides by DídacSurís) Abu-El-Haija, Sami, Nisarg Kothari, Joonseok Lee, Paul Natsev, George Toderici, Balakrishnan Varadarajan, and Sudheendra Vijayanarasimhan. "Youtube-8m: A large-scale video classification benchmark." arXiv preprint arXiv:1609.08675 (2016). [project] Activity Recognition: Datasets
  • 7.
    7 Recognition Figure: Tran, Du,Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
  • 8.
    8 Recognition Figure: Tran, Du,Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." ICCV 2015
  • 9.
    Karpathy, A., Toderici,G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. Large-scale video classification with convolutional neural networks. CVPR 2014 Slides extracted from ReadCV seminar by Victor Campos 9 Recognition: DeepVideo
  • 10.
    10 Recognition: DeepVideo: Demo Karpathy,A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video classification with convolutional neural networks. CVPR 2014
  • 11.
    11 Recognition: DeepVideo: Architectures Karpathy,A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video classification with convolutional neural networks. CVPR 2014
  • 12.
    12 Recognition: DeepVideo: Features Karpathy,A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video classification with convolutional neural networks. CVPR 2014
  • 13.
    13 Recognition: DeepVideo: Multiscale Karpathy,A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video classification with convolutional neural networks. CVPR 2014
  • 14.
    14 Recognition: DeepVideo: Results Karpathy,A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., & Fei-Fei, L. . Large-scale video classification with convolutional neural networks. CVPR 2014
  • 15.
    15 Jeffrey Donahue, LisaAnne Hendricks, Sergio Guadarrama, Marcus Rohrbach, Subhashini Venugopalan, Kate Saenko, Trevor Darrel. Long-term Recurrent Convolutional Networks for Visual Recognition and Description, CVPR 2015. code Activity Recognition: Frames + LSTM
  • 16.
    16 Yue-Hei Ng, Joe,Matthew Hausknecht, Sudheendra Vijayanarasimhan, Oriol Vinyals, Rajat Monga, and George Toderici. "Beyond short snippets: Deep networks for video classification." CVPR 2015 Activity Recognition: Frames + Optical Flow + LSTM
  • 17.
    17 Recognition Tran, Du, LubomirBourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." ICCV 2015
  • 18.
    18 Simonyan, Karen, andAndrew Zisserman. "Two-stream convolutional networks for action recognition in videos." NIPS 2014. Recognition: Two stream Two CNNs in paralel: ● One for RGB images ● One for Optical flow (hand-crafted features) Fusion after the softmax layer
  • 19.
    19Feichtenhofer, Christoph, AxelPinz, and Andrew Zisserman. "Convolutional two-stream network fusion for video action recognition." CVPR 2016. [code] Recognition: Two stream Two CNNs in paralel: ● One for RGB images ● One for Optical flow (hand-crafted features) Fusion at a convolutional layer
  • 20.
    20 Recognition Figure: Tran, Du,Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
  • 21.
    21 Recognition: C3D Figure: Tran,Du, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015
  • 22.
    22 Tran, Du, LubomirBourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015 Recognition: C3D: Demo
  • 23.
    23 Tran, Du, LubomirBourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015 Recognition: C3D: Temporal dimension 3D ConvNets are more suitable for spatiotemporal feature learning compared to 2D ConvNets Temporal depth 2D ConvNets
  • 24.
    24 Tran, Du, LubomirBourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015 A homogeneous architecture with small 3 × 3 × 3 convolution kernels in all layers is among the best performing architectures for 3D ConvNets Recognition: C3D: Temporal dimension
  • 25.
    25 Tran, Du, LubomirBourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015 Recognition: C3D: Architecture Feature vector
  • 26.
    26 Tran, Du, LubomirBourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 4489-4497. 2015 Recognition: C3D: Feature vector 16-frame clip 16-frame clip 16-frame clip 16-frame clip ... Average 4096-dimvideodescriptor 4096-dimvideodescriptor L2 norm
  • 27.
  • 28.
  • 29.
  • 30.
    30 Shou, Zheng, DongangWang, and Shih-Fu Chang. "Temporal action localization in untrimmed videos via multi-stage cnns." CVPR 2016. (Slidecast and Slides by Alberto Montes) Recognition: Localization
  • 31.
    31 Recognition: Localization (Slidecast andSlides by Alberto Montes) Shou, Zheng, Dongang Wang, and Shih-Fu Chang. "Temporal action localization in untrimmed videos via multi-stage cnns." CVPR 2016.
  • 32.
    32 Recognition: Localization (Slidecast andSlides by Alberto Montes) Shou, Zheng, Dongang Wang, and Shih-Fu Chang. "Temporal action localization in untrimmed videos via multi-stage cnns." CVPR 2016.
  • 33.
  • 34.
  • 35.
    Figure: Tran, Du,Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. "Learning spatiotemporal features with 3D convolutional networks." CVPR 2015 3D Convolutions over sets of 16 frames... 35 Temporal Activity Detection
  • 36.
  • 37.
    BSc thesis mAP = 0.5938mAP = 0.5492 mAP = 0.5635 Deeper networks present overfitting 37 Temporal Activity Detection
  • 38.
  • 39.
  • 40.
  • 41.
    BSc thesis Ground Truth: Playing waterpolo Prediction: 0.765 Playing water polo 0.202 Swimming 0.007 Springboard diving 41 Temporal Activity Detection
  • 42.
    BSc thesis Ground Truth: Hopscotch Prediction: 0.848 Runninga marathon 0.023 Triple jump 0.022 Javelin throw 42 Temporal Activity Detection
  • 43.
    BSc thesis 43 Temporal Activity Detection A.Montes, Salvador, A., Pascual-deLaPuente, S., and Giró-i-Nieto, X., “Temporal Activity Detection in Untrimmed Videos with Recurrent Neural Networks”, in 1st NIPS Workshop on Large Scale Computer Vision Systems 2016 (best poster award)
  • 44.
    44 Temporal Activity Detection A.Montes, Salvador, A., Pascual-deLaPuente, S., and Giró-i-Nieto, X., “Temporal Activity Detection in Untrimmed Videos with Recurrent Neural Networks”, in 1st NIPS Workshop on Large Scale Computer Vision Systems 2016 (best poster award)
  • 45.
    Outline 1. Recognition 2. OpticalFlow 3. Object Tracking 45
  • 46.
    Optical Flow Dosovitskiy, A.,Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 46
  • 47.
    Optical Flow: FlowNet Dosovitskiy,A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. ICCV 2015 47
  • 48.
    Optical Flow: FlowNet Dosovitskiy,A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 48 End to end supervised learning of optical flow.
  • 49.
    Optical Flow: FlowNet(contracting) Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 49 Option A: Stack both input images together and feed them through a generic network.
  • 50.
    Optical Flow: FlowNet(contracting) Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 50 Option B: Create two separate, yet identical processing streams for the two images and combine them at a later stage.
  • 51.
    Optical Flow: FlowNet(contracting) Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 51 Correlation layer: Convolution of data patches from the layers to combine. Option B: Create two separate, yet identical processing streams for the two images and combine them at a later stage.
  • 52.
    Optical Flow: FlowNet(expanding) Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 52 Upconvolutional layers: Unpooling features maps + convolution. Upconvolutioned feature maps are concatenated with the corresponding map from the contractive part.
  • 53.
    Optical Flow: FlowNet Dosovitskiy,A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. ICCV 2015 53 Since existing ground truth datasets are not sufficiently large to train a Convnet, a synthetic Flying Dataset is generated… and augmented (translation, rotation, scaling transformations; additive Gaussian noise; changes in brightness, contrast, gamma and color). Convnets trained on these unrealistic data generalize well to existing datasets such as Sintel and KITTI. Data augmentation
  • 54.
    Dosovitskiy, A., Fischer,P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., van der Smagt, P., Cremers, D. and Brox, T., 2015. FlowNet: Learning Optical Flow With Convolutional Networks. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2758-2766). 54 Optical Flow: FlowNet
  • 55.
    Outline 1. Recognition 2. OpticalFlow 3. Object Tracking 55
  • 56.
    Object tracking: Deepbut not CNN 56 Wang, Naiyan, and Dit-Yan Yeung. "Learning a deep compact image representation for visual tracking." NIPS 2013. Offline learning: Robust and generic features are learning by training a stacked denoising autoencoder on auxiliary images. Online learning: Encoder part of the autoencode + classification neural network
  • 57.
    Object tracking: MDNet 57 Nam,Hyeonseob, and Bohyung Han. "Learning multi-domain convolutional neural networks for visual tracking." ICCV VOT Workshop (2015)
  • 58.
    Object tracking: MDNet:Architecture 58 Nam, Hyeonseob, and Bohyung Han. "Learning multi-domain convolutional neural networks for visual tracking." ICCV VOT Workshop (2015) Domain-specific layers are used during training for each sequence, but are replaced by a single one at test time.
  • 59.
    Object tracking: MDNet:Online update 59 Nam, Hyeonseob, and Bohyung Han. "Learning multi-domain convolutional neural networks for visual tracking." ICCV VOT Workshop (2015) MDNet is updated online at test time with hard negative mining, that is, selecting negative samples with the highest positive score.
  • 60.
    Object tracking: FCNT 60 Wang,Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." ICCV 2015 [code]
  • 61.
    Object tracking: FCNT 61 Wang,Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." CVPR 2015 [code] Focus on conv4-3 and conv5-3 of VGG-16 network pre-trained for ImageNet image classification. conv4-3 conv5-3
  • 62.
    Object tracking: FCNT:Specialization 62 Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 3119-3127. 2015 [code] Most feature maps in VGG-16 conv4-3 and conv5-3 are not related to the foreground regions in a tracking sequence.
  • 63.
    Object tracking: FCNT:Localization 63 Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 3119-3127. 2015 [code] Although trained for image classification, feature maps in conv5-3 enable object localization… ...but is not discriminative enough to different objects of the same category.
  • 64.
    Object tracking: Localization 64 Zhou,Bolei, Aditya Khosla, Agata Lapedriza, Aude Oliva, and Antonio Torralba. "Object detectors emerge in deep scene cnns." ICLR 2015. Other works have shown how features maps in convolutional layers allow object localization.
  • 65.
    Object tracking: FCNT:Localization 65 Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 3119-3127. 2015 [code] On the other hand, feature maps from conv4-3 are more sensitive to intra-class appearance variation… conv4-3 conv5-3
  • 66.
    Object tracking: FCNT:Architecture 66 Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 3119-3127. 2015 [code] SNet=Specific Network (online update) GNet=General Network (fixed)
  • 67.
    Object tracking: FCNT:Results 67 Wang, Lijun, Wanli Ouyang, Xiaogang Wang, and Huchuan Lu. "Visual Tracking with Fully Convolutional Networks." In Proceedings of the IEEE International Conference on Computer Vision, pp. 3119-3127. 2015 [code]
  • 68.
    Object tracking: ROLO 68 Ning,Guanghan, Zhi Zhang, Chen Huang, Zhihai He, Xiaobo Ren, and Haohong Wang. "Spatially Supervised Recurrent Convolutional Neural Networks for Visual Object Tracking." IEEE International Symposium on Circuits and Systems, 2017
  • 69.
    Object tracking: ROLO 69 Ning,Guanghan, Zhi Zhang, Chen Huang, Zhihai He, Xiaobo Ren, and Haohong Wang. "Spatially Supervised Recurrent Convolutional Neural Networks for Visual Object Tracking." arXiv preprint arXiv:1607.05781 (2016)
  • 70.
    Object tracking: CFNet 70 Valmadre,Jack, Luca Bertinetto, João F. Henriques, Andrea Vedaldi, and Philip HS Torr. "End-to-end representation learning for Correlation Filter based tracking." CVPR 2017
  • 71.