RECORD
Deep Stereo Matching with Explicit Cost Aggregation Sub-Architecture
Computer Vision and Pattern Recognition · Submitted 12 Jan 2018
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim The comprehensive experiments on challenge datasets such as KITTI and Scene Flow show that our method outperforms the state-of-the-art methods.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Deep neural networks have shown excellent performance for stereo matching. Many efforts focus on the feature extraction and similarity measurement of the matching cost computation step while less attention is paid on cost aggregation which is crucial for stereo matching. In this paper, we present a learning-based cost aggregation method for stereo matching by a novel sub-architecture in the end-to-end trainable pipeline. We reformulate the cost aggregation as a learning process of the generation and selection of cost aggregation proposals which indicate the possible cost aggregation results. The cost aggregation sub-architecture is realized by a two-stream network: one for the generation of cost aggregation proposals, the other for the selection of the proposals. The criterion for the selection is determined by the low-level structure information obtained from a light convolutional network. The two-stream network offers a global view guidance for the cost aggregation to rectify the mismatching value stemming from the limited view of the matching cost computation. The comprehensive experiments on challenge datasets such as KITTI and Scene Flow show that our method outperforms the state-of-the-art methods.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Deep neural networks have shown excellent performance for stereo matching. | background 1.00 | 0.06 | 0.02 | asserted 1.00 |
| s1 | Many efforts focus on the feature extraction and similarity measurement of the matching cost computation step while less attention is paid on cost aggregation which is crucial for stereo matching. | gap 0.91 | 0.08 | 0.03 | asserted 1.00 |
| s2 | In this paper, we present a learning-based cost aggregation method for stereo matching by a novel sub-architecture in the end-to-end trainable pipeline. | contribution 1.00 | 0.93 | 0.11 | asserted 1.00 |
| s3 | We reformulate the cost aggregation as a learning process of the generation and selection of cost aggregation proposals which indicate the possible cost aggregation results. | method 0.80 | 0.91 | 0.06 | asserted 1.00 |
| s4 | The cost aggregation sub-architecture is realized by a two-stream network: one for the generation of cost aggregation proposals, the other for the selection of the proposals. | method 0.99 | 0.91 | 0.04 | asserted 1.00 |
| s5 | The criterion for the selection is determined by the low-level structure information obtained from a light convolutional network. | method 1.00 | 0.75 | 0.03 | asserted 1.00 |
| s6 | The two-stream network offers a global view guidance for the cost aggregation to rectify the mismatching value stemming from the limited view of the matching cost computation. | method 0.68 | 0.73 | 0.07 | asserted 1.00 |
| s7 | The comprehensive experiments on challenge datasets such as KITTI and Scene Flow show that our method outperforms the state-of-the-art methods. | result 1.00 | 0.96 | 0.81 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Deep neural networks have shown excellent performance for stereo matching. Many efforts focus on the feature extraction and similarity measurement of the matching cost computation step while less attention is paid on cost aggregation which is crucial for stereo matching. In this paper, we present a learning-based cost aggregation method for stereo matching by a novel sub-architecture in the end-to-end trainable pipeline. We reformulate the cost aggregation as a learning process of the generation and selection of cost aggregation proposals which indicate the possible cost aggregation results. The cost aggregation sub-architecture is realized by a two-stream network: one for the generation of cost aggregation proposals, the other for the selection of the proposals. The criterion for the selection is determined by the low-level structure information obtained from a light convolutional network. The two-stream network offers a global view guidance for the cost aggregation to rectify the mismatching value stemming from the limited view of the matching cost computation. The comprehensive experiments on challenge datasets such as KITTI and Scene Flow show that our method outperforms the state-of-the-art methods.
Human review
How the literature uses it
What this paper does with the arXiv papers it cites
What it cites · 1 sentence without a confident label
Uses the method or tool of 1 paper
-
Our architecture is implemented by the Tensoflow [ref] with a standard RMSProp [ref] and a constant learning rate of 0.0001.
TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems · Experimental Results · 0.99
Compares with 2 papers
-
In the Table [ref], the comparisons on KITTI2012 with deep stereo methods such as GC-net [ref] , Displets v2 [ref] , L-ResMatch [ref] and MC-CNN [ref] are shown, besides, the comparisons with other cost aggregation methods including CAT [ref] and S+GF [ref] are also illustrated in Table [ref], the leaderboard on KITTI2015 compares our method with GC-Net [ref] , MC-CNN [ref] , Displetv v2 [ref] , DRR [ref] , L-ResMatch [ref] and 3DMST [ref] .
Detect, Replace, Refine: Deep Structured Prediction For Pixel Wise Labeling · Benchmark results · 1.00
-
In the Table [ref], the comparisons on KITTI2012 with deep stereo methods such as GC-net [ref] , Displets v2 [ref] , L-ResMatch [ref] and MC-CNN [ref] are shown, besides, the comparisons with other cost aggregation methods including CAT [ref] and S+GF [ref] are also illustrated in Table [ref], the leaderboard on KITTI2015 compares our method with GC-Net [ref] , MC-CNN [ref] , Displetv v2 [ref] , DRR [ref] , L-ResMatch [ref] and 3DMST [ref] .
Improved Stereo Matching with Constant Highway Networks and Reflective Confidence Learning · Benchmark results · 1.00
Cites as background 1 paper
-
Many state-of-the-art methods raise the performance by learning robust local features or similarity measurements for cost computation [ref] .
Improved Stereo Matching with Constant Highway Networks and Reflective Confidence Learning · Introduction · 0.91
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.