diff --git a/src/datasets/AUSLAN.json b/src/datasets/AUSLAN.json index 7dad252..c568b52 100644 --- a/src/datasets/AUSLAN.json +++ b/src/datasets/AUSLAN.json @@ -1,15 +1,15 @@ { "pub": { "name": "AUSLAN", - "year": 2010, - "publication": "dataset:johnston2010archive", - "url": "https://elar.soas.ac.uk/Collection/MPI55247" + "year": 2008, + "publication": "dataset:johnston2008archive", + "url": "http://hdl.handle.net/2196/00-0000-0000-0000-D7CF-8" }, - "features": [], + "features": ["video", "gloss"], "language": "Australian", "#items": null, "#samples": "1,100 Videos", "#signers": 100, - "license": null, - "licenseUrl": null + "license": "Attribution", + "licenseUrl": "http://hdl.handle.net/2196/d8a991a5-d8cc-4f85-a5ff-c37279ebb625" } diff --git a/src/index.md b/src/index.md index 23500e5..73fe8b6 100644 --- a/src/index.md +++ b/src/index.md @@ -730,6 +730,11 @@ and so they have broken the dependency upon costly annotated gloss information i @shi-etal-2022-open introduce OpenASL, a large-scale American Sign Language (ASL) - English dataset collected from online video sites (e.g., YouTube), and then propose a set of techniques including sign search as a pretext task for pre-training and fusion of mouthing and handshape features to improve translation quality in the absence of glosses and in the presence of visually challenging data. +In the First WMT Shared Task [@muller-etal-2022-findings], they found that about half of the participants chose to epresent sign language data as video frames using a visual feature extractor on the encoder side. +All submitted systems were sequence-to-sequence models based on Transformers [@vaswani2017attention]. + + + @chen2022TwoStreamNetworkSign present a two-stream network for sign language recognition (SLR) and translation (SLT), utilizing a dual visual encoder architecture to encode RGB video frames and pose keypoints in separate streams. @@ -792,6 +797,10 @@ and showed similar performance, with the transformer underperforming on the vali They experimented with various normalization schemes, mainly subtracting the mean and dividing by the standard deviation of every individual keypoint either concerning the entire frame or the relevant "object" (Body, Face, and Hand). +In the First WMT Shared Task [@muller-etal-2022-findings], the baseline system [@mueller2022sign-sockeye-baselines] used pose inputs. +In addition, they found that about half of the participants [@tarres-etal-2022-tackling;@hufe-avramidis-2022-experimental] chose to represent signed language data as poses. +All submitted systems were sequence-to-sequence models based on Transformers [@vaswani2017attention]. + #### Text-to-Pose Text-to-Pose, also known as sign language production, is the task of producing a sequence of poses that adequately represent a spoken language text in sign language, as an intermediate representation to overcome challenges in animation. diff --git a/src/references.bib b/src/references.bib index f01e75d..6f5fe76 100644 --- a/src/references.bib +++ b/src/references.bib @@ -496,7 +496,7 @@ @article{dataset:schembri2013building year = {2013} } -@inproceedings{dataset:johnston2010archive, +@inproceedings{dataset:johnston2008archive, address = {The University of the Philippines Visayas Cebu College, Cebu City, Philippines}, author = {Johnston, Trevor}, booktitle = {Proceedings of the 22nd Pacific Asia Conference on Language, Information and Computation}, @@ -2352,6 +2352,82 @@ @inproceedings{muller-etal-2022-findings year = {2022} } +@inproceedings{hufe-avramidis-2022-experimental, + title = "Experimental Machine Translation of the {S}wiss {G}erman Sign Language via 3{D} Augmentation of Body Keypoints", + author = "Hufe, Lorenz and + Avramidis, Eleftherios", + booktitle = "Proceedings of the Seventh Conference on Machine Translation (WMT)", + month = dec, + year = "2022", + publisher = "Association for Computational Linguistics", + url = "https://aclanthology.org/2022.wmt-1.95", + pages = "983--988" +} + +@inproceedings{tarres-etal-2022-tackling, + title = "Tackling Low-Resourced Sign Language Translation: {UPC} at {WMT}-{SLT} 22", + author = "Tarres, Laia and + G{\'a}llego, Gerard I. and + Giro-i-nieto, Xavier and + Torres, Jordi", + booktitle = "Proceedings of the Seventh Conference on Machine Translation (WMT)", + month = dec, + year = "2022", + publisher = "Association for Computational Linguistics", + url = "https://aclanthology.org/2022.wmt-1.97", + pages = "994--1000" +} + +@inproceedings{hamidullah-etal-2022-spatio, + title = "Spatio-temporal Sign Language Representation and Translation", + author = "Hamidullah, Yasser and + Van Genabith, Josef and + Espa{\~n}a-bonet, Cristina", + booktitle = "Proceedings of the Seventh Conference on Machine Translation (WMT)", + month = dec, + year = "2022", + publisher = "Association for Computational Linguistics", + url = "https://aclanthology.org/2022.wmt-1.94", + pages = "977--982" +} + +@inproceedings{dey-etal-2022-clean, + title = "Clean Text and Full-Body Transformer: {M}icrosoft{'}s Submission to the {WMT}22 Shared Task on Sign Language Translation", + author = "Dey, Subhadeep and + Pal, Abhilash and + Chaabani, Cyrine and + Koller, Oscar", + booktitle = "Proceedings of the Seventh Conference on Machine Translation (WMT)", + month = dec, + year = "2022", + publisher = "Association for Computational Linguistics", + url = "https://aclanthology.org/2022.wmt-1.93", + pages = "969--976" +} + +@inproceedings{shi-etal-2022-ttics, + title = "{TTIC}{'}s {WMT}-{SLT} 22 Sign Language Translation System", + author = "Shi, Bowen and + Brentari, Diane and + Shakhnarovich, Gregory and + Livescu, Karen", + booktitle = "Proceedings of the Seventh Conference on Machine Translation (WMT)", + month = dec, + year = "2022", + publisher = "Association for Computational Linguistics", + url = "https://aclanthology.org/2022.wmt-1.96", + pages = "989--993" +} + +@misc{mueller2022sign-sockeye-baselines, + title={Sockeye baseline models for sign language translation}, + author={M\"{u}ller, Mathias and Rios, Annette and Moryossef, Amit}, + howpublished={\url{https://github.com/bricksdont/sign-sockeye-baselines}}, + year={2022} +} + + + @inproceedings{shi-etal-2022-open, address = {Abu Dhabi, United Arab Emirates}, author = {Shi, Bowen and