Large-scale capture of hidden fluorescent labels for training generalizable markerless motion capture models.

Deep learning-based markerless tracking has revolutionized studies of animal behavior. Yet the generalizability of trained models tends to be limited, as new training data typically needs to be generated manually for each setup or visual environment. With each model trained from scratch, researchers track distinct landmarks and analyze the resulting kinematic data in idiosyncratic ways. Moreover, due to inherent limitations in manual annotation, only a sparse set of landmarks are typically labeled. To address these issues, we developed an approach, which we term GlowTrack, for generating orders of magnitude more training data, enabling models that generalize across experimental contexts. We describe: a) a high-throughput approach for producing hidden labels using fluorescent markers; b) a multi-camera, multi-light setup for simulating diverse visual conditions; and c) a technique for labeling many landmarks in parallel, enabling dense tracking. These advances lay a foundation for standardized behavioral pipelines and more complete scrutiny of movement.

Impact of correcting visual impairment and low vision in deaf-mute students in Pune, India.

AIM: The aim of this study was to evaluate visual acuity and vision function before and after providing spectacles and low vision devices (LVDs) in deaf-mute students. SETTINGS: Schools for deaf-mute in West Maharashtra. METHODS: Hearing-impaired children in all special schools in Pune district underwent detailed visual acuity testing (with teachers' help), refraction, external ocular examination, and fundoscopy. Students with refractive errors and low vision were provided with spectacles and LVD. The LV Prasad-Functional Vision Questionnaire consisting of twenty items was administered to each subject before and after providing spectacles, LVDs. STATISTICAL ANALYSIS: Wilcoxon matched-pairs signed-ranks test. RESULTS: 252/929 (27.1%) students had a refractive error. 794 (85.5%) were profound deaf. Two-hundred and fifty students were dispensed spectacles and LVDs. Mean LogMAR visual acuity before introduction of spectacles and LVDs were 0.33 ± 0.36 which improved to 0.058 (P < 0.0001) after intervention. It was found that difference in functional vision pre- and post-intervention was statistically significant (P < 0.0001) for questions 1-19. The most commonly reported difficulties were for performing distance task like reading the bus destination (58.7%), making out the bus number (51.1%), copying from blackboard (47.7%), and seeing whether somebody is waving hand from across the road (45.5%). In response to question number 20, 57.4% of students felt that their vision was much worse than their friend's vision, which was reduced to 17.6% after dispensing spectacles and LVDs. CONCLUSION: Spectacle and LVD reduced visual impairment and improved vision function in deaf-mute students, augmenting their ability to negotiate in and out of school.