A Study on the 4D Sparsity of JPEG Pleno Light Fields Using the Discrete Cosine Transform

Alves, Gustavo; Pereira, Marcio P.; Carvalho, Murilo B. de; Pereira, Fernando; Pagliari, Carla L.; Testoni, Vanessa; Silva, Eduardo A. B. da

doi:10.1109/icip.2018.8451583

Cited by 9 publications

(13 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…3 leads to the conclusion that for the lenslet LFs Bikes and Fountain, the energy decay is much slower along the (u, v) coordinates than along the (s, t) coordinates. Since the faster the 4D-DCT coefficients energy decay along a direction, the larger the redundancy along this direction, these results are coherent with the findings in [31], which state that for lenslet LFs the inter-view redundancy tends to be larger than the intraview redundancy. On the other hand, for the more sparsely angular sampled LFs Greek and Set2 2K sub, the energy decay along the (s, t) coordinates tends to be much slower than for the lenslet LFs.…”

Section: ) 4d-dct Coefficients Energy Distribution For Densely Angulsupporting

confidence: 87%

“…This is equivalent to saying that the 4D-DCT energy compaction properties are poorer for such LFs, which is mainly due to the larger baseline. These results are again coherent with the findings in [31]. Such poor energy compaction property of the 4D-DCT for the LFs which angular sampling is more sparse (non-lenslet) is the main reason why the JPEG Pleno 4D-Transform coding mode [37] performs better for lenslet LFs.…”

Section: ) 4d-dct Coefficients Energy Distribution For Densely Angulsupporting

confidence: 85%

“…In [29], the authors employ compressive sensing and convolutional neural networks to code a LF as a 4D tensor. Studies on the sparsity of the 4D-DCT coefficients associated to a LF [30], [31] led to the development of a simple LF codec based on the separable 4D-DCT [32], referred to as MuLE (Multidimensional Light Field Encoder); this codec shows good RD performance when there is large inter-view redundancy as for lenslet LFs and it is an early version of the LF codec taken as the main focus of this paper, the JPEG Pleno Light Field Coding 4D-Transform mode.…”

Section: A Most Relevant Light Field Coding Solutionsmentioning

confidence: 99%

“…Densely angular sampled LFs tend to show a large amount of spatio-angular redundancy [31]. When applying the 4D-DCT to a 4D block, it is expected that most of the data block energy is concentrated in the lower frequency transform coefficients.…”

Section: ) 4d-dct Coefficients Energy Distribution For Densely Angulmentioning

confidence: 99%

See 3 more Smart Citations

The JPEG Pleno Light Field Coding Standard 4D-Transform Mode: How to Design an Efficient 4D-Native Codec

et al. 2020

Self Cite

View full text Add to dashboard Cite

The increasing demand for highly realistic and immersive visual experiences has led to the emergence of richer 3D visual representation models such as light fields, point clouds and meshes. Light fields may be modelled as a 2D array of 2D views, corresponding to a large amount of data, which demands for highly efficient coding solutions. Although static light fields are inherently 4D structures, the coding solutions in the literature mostly employ traditional 2D coding tools associated with techniques such as depth-based image rendering to generate a residual, again to be coded using available 2D coding tools. To address the market needs, JPEG has launched the so-called JPEG Pleno standard, which Part 2 is dedicated to light field coding. The JPEG Pleno Light Field Coding standard includes two coding modes, one based on the 4D-DCT, so-called 4D-Transform mode, and another based on depth-based synthesis, so-called 4D-Prediction. The 4D-Transform coding mode standardizes for the first time a 4D-native light field coding solution where the full light field redundancy across the four dimensions is comprehensively exploited, somehow extending to 4D the 2D coding framework adopted decades ago by the popular JPEG Baseline standard. In this context, this paper describes and analyzes in detail the conceptual and algorithmic design process which has led to the creation of the JPEG Pleno Light Field Coding standard 4D-Transform coding mode. This has happened through a sequence of steps involving technical innovation design and integration where increasingly sophisticated coding tools have been combined and improved to maximize the final rate distortion (RD) performance. INDEX TERMS JPEG Pleno standard, Light field coding, 4D-Transform mode, 4D-DCT.

show abstract

Section: ) 4d-dct Coefficients Energy Distribution For Densely Angulsupporting

confidence: 87%

Section: ) 4d-dct Coefficients Energy Distribution For Densely Angulsupporting

confidence: 85%

Section: A Most Relevant Light Field Coding Solutionsmentioning

confidence: 99%

Section: ) 4d-dct Coefficients Energy Distribution For Densely Angulmentioning

confidence: 99%

See 2 more Smart Citations

The JPEG Pleno Light Field Coding Standard 4D-Transform Mode: How to Design an Efficient 4D-Native Codec

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Porém, a representação mais natural de um light field parece ser por meio de uma estrutura 4-dimensional, com duas dimensões para o plano das vistas e outras duas para o plano espacial da imagem. Tal possibilidadeé considerada em [11], em que se demonstra que, usando a 4D-DCT, conseguese explorar com maior eficiência a redundância presente na estrutura de um light field lenslet do que usando suas versões bidimensionais. Essa conclusão deu suporte para que, em [12], fosse proposto um CODEC para imagens lenslet de light fields utilizando a 4D-DCT, explorando a alta correlação entre as vistas e o plano espacial.…”

Section: Introductionunclassified

Um Esquema para Inserção de Marcas D'Água em Imagens de Light Field

Ferreira¹,

Lima²

2019

Anais De XXXVII Simpósio Brasileiro De Telecomunicações E Processamento De Sinais

View full text Add to dashboard Cite

Resumo-Imagens produzidas a partir de light fields representam um avanço recente e importante em diversas aplicações. Porém, tais imagens trazem também novos desafios relacionados, por exemplo,à transmissão eà segurança dos dados. Esteúltimo aspectoé abordado no presente trabalho, em que se propõe um esquema cego e baseado em Quantization Index Modulation para inserção de marcas d'água em imagens lenslet de light field. O objetivoé recuperar a marca, mesmo após a imagem ter sido submetidaà compressão no domínio da transformada discreta do cosseno em quatro dimensões; para isso, desenvolve-se uma estratégia de ordenação dos coeficientes no referido domínio. Os testes realizados demonstram que a marca proposta atende aos requisitos de imperceptibilidade e robustez desejados. Palavras-Chave-Marca d'água, Quantization Index Modulation, Light field, transformada discreta do cosseno.

show abstract

Watermarking and Coefficient Scanning for Light Field Images in 4D-DCT Domain

Ferreira¹,

Lima²

2021

IEEE Access

View full text Add to dashboard Cite

A Study on the 4D Sparsity of JPEG Pleno Light Fields Using the Discrete Cosine Transform

Cited by 9 publications

References 3 publications

The JPEG Pleno Light Field Coding Standard 4D-Transform Mode: How to Design an Efficient 4D-Native Codec

The JPEG Pleno Light Field Coding Standard 4D-Transform Mode: How to Design an Efficient 4D-Native Codec

Um Esquema para Inserção de Marcas D'Água em Imagens de Light Field

Watermarking and Coefficient Scanning for Light Field Images in 4D-DCT Domain

Contact Info

Product

Resources

About