Multi-image super-resolution using a dual weighting scheme based on Voronoi tessellation

Bätz, Michel; Eichenseer, Andrea; Kaup, André

doi:10.1109/icip.2016.7532874

Cited by 10 publications

(12 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For SISR, we study dictionary and deep learning methods. The dictionary methods are example-based ridge regression (EBSR) [9], sparse coding (ScSR) [77], Naive Bayes SR Interpolation BICUBIC NUISR [74], WNUISR [18] DBRSR [75] Non-blind L1BTV [22], BEPSR [78] reconstruction IRWSR [23] Blind reconstruction BVSR [24], SRB [62] forests (NBSRF) [10], and adjusted anchored neighborhood regression (A+) [12]. The deep learning methods comprise CNNs (SRCNN) [13], very deep networks (VDSR) [14], and deeply-recursive networks (DRCN) [15].…”

Section: Evaluated Algorithmsmentioning

confidence: 99%

“…Here, the sparsity priors contributed to the recovery of the printed text. (a) LR input (b) A+ [12] (c) VDSR [14] (d) WNUISR [18] (e) IRWSR [23] (f) Ground truth Fig. 7: SR methods under global motion on the newspapers dataset (3× magnification).…”

Section: Super-resolution On Static Scenesmentioning

confidence: 99%

“…We found that algorithms building on simple interpolation (NUISR, HYSR) were most sensitive. Interpolation-based SR with proper outlier (a) A+ [12] (b) NUISR [18] (c) IRWSR [23] (d) VSRnet [25] (e) Ground truth Fig. 9: SR on the dynamic coffee dataset (3× magnification).…”

Section: Super-resolution Under Video Compressionmentioning

confidence: 99%

“…The interpolation-based NUISR and WNUISR are slightly faster but achieve lower image quality. However, interpolation-based SR doubles the complexity of motion estimation as optical flow is computed in forward and backward direction [18].…”

Section: Image Quality and Computation Time Tradeoffmentioning

confidence: 99%

“…This explains the weaker performance on mixed or local motion as well as photometric variations, while SISR is unaffected by these factors. Within MFSR, reconstruction [22], [23], [78] or deep learning [25] methods are more robust than interpolation-based approaches [18], [74], [75].…”

Section: Remarks On the Quantitative Studymentioning

confidence: 99%

See 4 more Smart Citations

Toward Bridging the Simulated-to-Real Gap: Benchmarking Super-Resolution on Real Data

Köhler

Bätz

Naderi

et al. 2019

IEEE Trans. Pattern Anal. Mach. Intell.

Self Cite

View full text Add to dashboard Cite

Capturing ground truth data to benchmark super-resolution (SR) is challenging. Therefore, current quantitative studies are mainly evaluated on simulated data artificially sampled from ground truth images. We argue that such evaluations overestimate the actual performance of SR methods compared to their behavior on real images. Toward bridging this simulated-to-real gap, we introduce the Super-Resolution Erlangen (SupER) database, the first comprehensive laboratory SR database of all-real acquisitions with pixel-wise ground truth. It consists of more than 80k images of 14 scenes combining different facets: CMOS sensor noise, real sampling at four resolution levels, nine scene motion types, two photometric conditions, and lossy video coding at five levels. As such, the database exceeds existing benchmarks by an order of magnitude in quality and quantity. This paper also benchmarks 19 popular single-image and multi-frame algorithms on our data. The benchmark comprises a quantitative study by exploiting ground truth data and qualitative evaluations in a large-scale observer study. We also rigorously investigate agreements between both evaluations from a statistical perspective. One interesting result is that top-performing methods on simulated data may be surpassed by others on real data. Our insights can spur further algorithm development, and the publicy available dataset can foster future evaluations.

show abstract