Which Is Which? Evaluation of Local Descriptors for Image Matching in Real-World Scenarios

Bellavia, Fabio; Colombo, Carlo

doi:10.1007/978-3-030-29888-3_24

Cited by 4 publications

(9 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Heinly [42] and Oxford [13] datasets are other benchmarks which evaluate the descriptor performance under changes in viewpoint, scale, illumination, blur, rotation and compression. WISW benchmark [28] includes 5 sequences with changes in viewpoint. The features were assessed by mean average precision (mAP).…”

Section: Homography Datasetsmentioning

confidence: 99%

Interwoven texture-based description of interest points in images

Ghahremani

Zhao

Tiddeman

et al. 2021

Pattern Recognition

View full text Add to dashboard Cite

Local feature description is to assign a unique signature to a key-point such that it becomes distinctive from the others regardless of changes in viewpoint, illumination, rotation, scale as well as distortions and noise. This paper proposes a novel approach to construct such a descriptor. For preserving both homogeneous and heterogeneous features of a given support region, we interweave the texture information so that the key-point is more likely to be assigned a distinctive signature and neighboring key-points will be less likely to share the same texture information.The main idea behind our descriptor is to increase the areas of our observations in the given scene while the length of the local support region is fixed. Gradient magnitude and divergence, as measurement parameters of texture information, are applied to a group of pixels instead of employing a pixel-wise strategy that make the descriptor more resistant to noise, distortions and illumination variation. The required storage of the proposed descriptor is just 72 floats and its computational complexity is much lower than those of existing ones. A comparative study between the proposed method and the selected state-of-the-art ones over multiple publicly accessible datasets with different characteristics shows its superiority, robustness and computational efficiency under various geometric changes, illumination variation, distortions and noise.

show abstract

Section: Homography Datasetsmentioning

confidence: 99%

Interwoven texture-based description of interest points in images

Ghahremani

Zhao

Tiddeman

et al. 2021

Pattern Recognition

View full text Add to dashboard Cite

show abstract

“…As the Hamming distance H between two of these unpacked vectors is the same as the L 1 distance between the original vectors, this representation"trick" can save matching time on all hardware configurations where H is faster than L 1 to compute. Table 1 reports on image matching results obtained with the proposed quantization scheme with planar scenes from the WISW [10] and Oxford [44] datasets (19 sequences in all, 15 from WISW and 8 from Oxford, 4 sequences being in common between the datasets. Each sequence contains 6 images, the first of which is used as reference, for a total of 19 × (6 − 1) = 95 image pairs) which include viewpoint changes, the most challenging sources of image distortion, also in combinations with other relevant image transformations, such as illumination changes.…”

Section: Packed Siftmentioning

confidence: 99%

“…In the following, a new and general benchmark for non-planar scenes is introduced, that extends and refines the one proposed in [11] and employed in a recent descriptor evaluation [10]. This benchmark is aimed at evaluating descriptor behavior in real-world scenes, thus providing a deep insight into descriptor characteristics, while compensating for the limitations of both the planar and the application-based benchmarks.…”

Section: Related Workmentioning

confidence: 99%

“…5.2. The same dataset was also used for the WISW contest [10], but with a less accurate method for overlap error estimation than the one used here. The question arises whether the precision of the proposed descriptor matching evaluation depends on the thresholds values t ε (maximum allowed approximated overlap error) and t X (patch center distance).…”

Section: Non-planar Scenesmentioning

confidence: 99%

“…The HPatches dataset uses square patches that cannot be rotated using patch orientation assignment without losing patch data[10]. As this would compromise the fairness of the comparison, HPatches is not considered in the present evaluation.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Is There Anything New to Say About SIFT Matching?

Bellavia

Colombo

2020

Int J Comput Vis

Self Cite

View full text Add to dashboard Cite

SIFT is a classical hand-crafted, histogrambased descriptor that has deeply affected research on image matching for more than a decade. In this paper, a critical review of the aspects that affect SIFT matching performance is carried out, and novel descriptor design strategies are introduced and individually evaluated. These encompass quantization, binarization and hierarchical cascade filtering as means to reduce data storage and increase matching efficiency, with no significant loss of accuracy. An original contextual matching strategy based on a symmetrical variant of the usual nearest-neighbor ratio is discussed as well, that can increase the discriminative power of any descriptor. The paper then undertakes a comprehensive experimental evaluation of state-of-the-art hand-crafted and datadriven descriptors, also including the most recent deep descriptors. Comparisons are carried out according to several performance parameters, among which accuracy and space-time efficiency. Results are provided for both planar and non-planar scenes, the latter being evaluated with a new benchmark based on the concept of approximated patch overlap. Experimental evidence shows that, despite their age, SIFT and other handcrafted descriptors, once enhanced through the proposed strategies, are ready to meet the future image matching challenges. We also believe that the lessons learned from this work will inspire the design of better hand-crafted and data-driven descriptors.

show abstract

RootsGLOH2: embedding RootSIFT ‘square rooting’ in sGLOH2

Bellavia¹,

Colombo²

2020

IET Computer Vision

Self Cite

View full text Add to dashboard Cite

This paper introduces an extension of the sGLOH2 local image descriptor inspired by RootSIFT "square rooting" as a way to indirectly alter the matching distance used to compare the descriptor vectors. The extended descriptor, named Roots-GLOH2, achieved the best results in terms of matching accuracy and robustness among the latest state-of-the-art non-deep descriptors in recent evaluation contests dealing with both planar and non-planar scenes. RootsGLOH2 also achieves a matching accuracy very close to that obtained by the best deep descriptors to date. Beside confirming that "square rooting" has beneficial effects on sGLOH2 as it happens on SIFT, experimental evidence shows that classical norm-based distances, such as the Euclidean and Manhattan distances, only provide suboptimal solutions to the problem of local image descriptor matching. This suggests matching distance design as a topic to investigate further in the near future.

show abstract

Which Is Which? Evaluation of Local Descriptors for Image Matching in Real-World Scenarios

Cited by 4 publications

References 37 publications

Interwoven texture-based description of interest points in images

Interwoven texture-based description of interest points in images

Is There Anything New to Say About SIFT Matching?

RootsGLOH2: embedding RootSIFT ‘square rooting’ in sGLOH2

Contact Info

Product

Resources

About