The 9th annual MLSP competition: New methods for acoustic classification of multiple simultaneous bird species in a noisy environment

Briggs, Forrest; Huang, Yonghong; Raich, Raviv; Eftaxias, K.; Zhao, Lei; Cukierski, William; Hadley, Sarah Frey; Hadley, Adam S.; Betts, Matthew G.; Fern, Xiaoli Z.; Irvine, Jed; Neal, Lawrence; Thomas, Anil; Fodor, Gábor; Tsoumakas, Grigorios; Ng, Hong Wei; Nguyen, Thi Ngoc Tho; Huttunen, Heikki; Ruusuvuori, Pekka; Manninen, Tapio; Diment, Aleksandr; Virtanen, Tuomas; Marzat, Julien; Defretin, Joseph; Callender, Dave; Hurlburt, Chris; Larrey, Ken; Milakov, Maxim

doi:10.1109/mlsp.2013.6661934

Cited by 70 publications

(43 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Before LifeCLEF started in 2014, three previous initiatives on the evaluation of acoustic bird species identification took place, including two from the SABIOD 15 group [25,24,7]. In collaboration with the organizers of these previous challenges, BirdCLEF 2014, 2015 and 2016 challenges went one step further by (i) significantly increasing the species number by an order of magnitude, (ii) working on real-world social data built from thousands of recordists, and (iii) moving to a more usage-driven and system-oriented benchmark by allowing the use of meta-data and defining information retrieval oriented metrics.…”

Section: Task2: Birdclefmentioning

confidence: 99%

LifeCLEF 2015: Multimedia Life Species Identification Challenges

Joly

Goëau

Glotin

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Using multimedia identification tools is considered as one of the most promising solutions to help bridge the taxonomic gap and build accurate knowledge of the identity, the geographic distribution and the evolution of living species. Large and structured communities of nature observers (e.g., iSpot, Xeno-canto, Tela Botanica, etc.) as well as big monitoring equipment have actually started to produce outstanding collections of multimedia records. Unfortunately, the performance of the state-of-the-art analysis techniques on such data is still not well understood and is far from reaching real world requirements. The LifeCLEF lab proposes to evaluate these challenges around 3 tasks related to multimedia information retrieval and fine-grained classification problems in 3 domains. Each task is based on large volumes of real-world data and the measured challenges are defined in collaboration with biologists and environmental stakeholders to reflect realistic usage scenarios. For each task, we report the methodology, the data sets as well as the results and the main outcomes.

show abstract

Section: Task2: Birdclefmentioning

confidence: 99%

LifeCLEF 2015: Multimedia Life Species Identification Challenges

Joly

Goëau

Glotin

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…This work builds upon certain publications (Briggs et al, 2012(Briggs et al, , 2013a(Briggs et al, , 2013bFodor, 2013;Kridler, 2013;Lasseck, 2013Lasseck, , 2014Potamitis, 2014) and continues their line of thought. It will focus around a core process: The unsupervised extraction of templates from spectrograms and their cataloguing as to become a dictionary of spectral patches belonging to vocalising species.…”

Section: Introductionmentioning

confidence: 57%

“…Bird recognition as a multi-label classification task (Briggs et al, 2012(Briggs et al, , 2013a If we treat the S matrix as a multi-label problem then we can apply the one-vs-all strategy for obtaining per species probabilities. If the codebook is composed of a total of N ROIs then each ROI is crosscorrelated with every recording of the dataset but only within the frequency boundaries of each ROI.…”

Section: Bird Recognition Cast As a Multi-label Classification Or Regmentioning

confidence: 99%

“…This work wants to analyse solely the unsupervised extraction of ROIs and their discriminative power of sweeping and classification process. In Briggs et al (2012Briggs et al ( , 2013aBriggs et al ( , 2013b, Fodor (2013), Keen et al (2014), Kridler (2013), Lasseck (2013Lasseck ( , 2014, Potamitis (2014), and Stowell and Plumbley (2014) other features are also integrated that focus on the shape of each segment or the marginal across time statistical properties of the spectrogram.…”

Section: Bird Recognition Cast As a Multi-label Classification Or Regmentioning

confidence: 99%

“…We show how the process of cross-correlating the codebook with the recordings returns a single matrix that can be analysed with two different ways: either as a multi-label problem (Briggs et al, 2012(Briggs et al, , 2013a or as a regression problem (Fodor, 2013;Lasseck, 2013Lasseck, , 2014. The reason we explore further this school of thought is that the multiple templates approach -in its numerous variations -was a building block for entries that scored first place in five very recent and very competitive bioacoustics recognition challenges (see (Bas et al, 2013;Briggs et al, 2013a) and Appendix A), a fact that cannot be disregarded. Therefore, to our point of view, and although it is too early to state this with confidence, it has the potential to become a benchmark for classifying bioacoustics data.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations