2014
DOI: 10.1007/978-3-319-10593-2_31
|View full text |Cite
|
Sign up to set email alerts
|

Jointly Optimizing 3D Model Fitting and Fine-Grained Classification

Abstract: Abstract. 3D object modeling and fine-grained classification are often treated as separate tasks. We propose to optimize 3D model fitting and fine-grained classification jointly. Detailed 3D object representations encode more information (e.g., precise part locations and viewpoint) than traditional 2D-based approaches, and can therefore improve fine-grained classification performance. Meanwhile, the predicted class label can also improve 3D model fitting accuracy, e.g., by providing more detailed classspecific… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
100
0
1

Year Published

2016
2016
2020
2020

Publication Types

Select...
5
3
2

Relationship

0
10

Authors

Journals

citations
Cited by 102 publications
(103 citation statements)
references
References 23 publications
2
100
0
1
Order By: Relevance
“…Taniai Benchmark [45] We first evaluated our FCSS descriptor on the Taniai benchmark [45], which consists of 400 image pairs divided into three groups: FG3DCar [29], JODS [37], and PASCAL [20]. As in [45], flow accuracy was measured by computing the proportion of foreground Figure 6.…”
Section: Resultsmentioning
confidence: 99%
“…Taniai Benchmark [45] We first evaluated our FCSS descriptor on the Taniai benchmark [45], which consists of 400 image pairs divided into three groups: FG3DCar [29], JODS [37], and PASCAL [20]. As in [45], flow accuracy was measured by computing the proportion of foreground Figure 6.…”
Section: Resultsmentioning
confidence: 99%
“…However, the results of Zia et al (2013) show that their approach heavily depends on a good pose initialisation. Similarly, Lin et al (2014) recover the 3D vehicle geometry by fitting the 3D ASM to estimated 2D landmark locations resulting from a DPM detector. Their approach also suffers from wrongly estimated part locations resulting from the DPM.…”
Section: Related Workmentioning
confidence: 99%
“…Top: A series of prior 3D shape basis [2]. Bottom: The shape estimation procedure for a given input image.…”
Section: Input Image Landmarks Localisationmentioning
confidence: 99%