Make3D
Convert your still image into 3D model: State-of-the-art results.
The Table below shows the state-of-the-art results of the depth prediction
on the state-of-the-art algorithms.
(This is on Make3D Range Image Dataset-1.)
| Algorithm | Main contribution | Relative Error | Log-10 error | Qualitative | RMSE |
|---|---|---|---|---|---|
| Data baseline. | Predict Average | 0.698 | 0.334 | 0% | (coming soon) |
| Saxena, Sung, Ng, NIPS 2005. [1,2] | Learn Depth: MRF | 0.530 | 0.198 | NR | 16.7m |
| Saxena, Sun, Ng, 3dRR/PAMI, 2007. [3,4] | Pointwise MRF | 0.458 | 0.149 | 23.0% | NR |
| Saxena, Sun, Ng, 3dRR/PAMI 2007. [3,4] | Superpixel MRF | 0.370 | 0.187 | 64.9% | NR |
| Hoiem, Efros, Hebert, IJCV 2007 [5] | Surface Layout | 1.423 | 0.320 | 33.1% | NR |
| Heitz, Gould, Saxena, Koller, NIPS 2008. [6] | Cascaded Models | NR | NR | NR | 15.4m |
| Cherian, Morellas, Papanikolopoulos, ICRA 2009. | Ground plane | NR | NR | NR | 22m |
| Liu, Gould, Koller, CVPR 2010. [7] | Semantic Labels | 0.375 | 0.148 | NR | NR |
| Li, Kowdle, Saxena, Chen, NIPS 2010. [8,9,10] | Feedback Cascades | NR | NR | NR | 15.2m |
| Li, Saxena, Chen, NIPS 2011. [11] | theta-MRF | NR | NR | NR | 15.0m |
| Karsch, Liu, Kang, ECCV 2012. [12] | depth transfer | 0.361 | 0.148 | NR | 15.1m |
NR - Not Reported.
References
-
Learning Depth from Single Monocular Images,
Ashutosh Saxena, Sung H. Chung, Andrew Y. Ng. In Neural Information Processing Systems (NIPS) 18, 2005. [pdf] -
3-D Depth Reconstruction from a Single Still Image,
Ashutosh Saxena, Sung H. Chung, Andrew Y. Ng. International Journal of Computer Vision (IJCV), Aug 2007. [pdf] -
Make3D: Learning 3-D Scene Structure from a Single Still Image,
Ashutosh Saxena, Min Sun, Andrew Y. Ng, In IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2008. [pdf] -
Learning 3-D Scene Structure from a Single Still Image,
Ashutosh Saxena, Min Sun, Andrew Y. Ng, In ICCV workshop on 3D Representation for Recognition (3dRR-07), 2007. (Best paper award.) [pdf, ppt, Talk] -
Recovering Surface Layout from an Image.
D. Hoiem, A.A. Efros, and M. Hebert. IJCV 2007. -
Cascaded Classification Models: Combining Models for Holistic Scene Understanding.
Geremy Heitz, Stephen Gould, Ashutosh Saxena, Daphne Koller. In Neural Information Processing Systems (NIPS), 2008. [pdf, more] -
Single Image Depth Estimation from Predicted Semantic Labels.
Beyang Liu, Stephen Gould and Daphne Koller. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2010. -
A Generic Model to Compose Vision Modules for Holistic Scene Understanding.
Adarsh Kowdle, Congcong Li, Ashutosh Saxena and Tsuhan Chen. In European Conference on Computer Vision Workshop on Parts and Attributes (ECCV '10), 2010. -
Towards Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models.
Congcong Li, Adarsh Kowdle, Ashutosh Saxena, Tsuhan Chen. In Neural Information Processing Systems (NIPS), 2010. -
Towards Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models.
Congcong Li, Adarsh Kowdle, Ashutosh Saxena, Tsuhan Chen. To appear in IEEE Trans Pattern Analysis and Machine Intelligence (PAMI), 2011. -
θ-MRF: Capturing Spatial and Semantic Structure in the Parameters for Scene Understanding, Congcong Li, Ashutosh Saxena, Tsuhan Chen. In Neural Information Processing Systems (NIPS), 2011.
Depth Extraction from Video Using Non-parametric Sampling, Kevin Karsch, Ce Liu, and Sing Bing Kang. In European Conference on Computer Vision (ECCV), 2012.