Tag: Dogs

Posted on

Wearables for Sled Dogs

On the last day of the ACI Conference (December 8, 2022), “Session 5: Sensors & Signals, Part II: Electric Boogaloo” started after the lunch break. Carlos Alberto Aguilar-Lazcano (CICESE-UT3) gave a talk on the topic “Towards a monitoring and emergency alarm system activated by the barking of assistant dogs”. The next presentation was “WAG’D: Towards a Wearable Activity and Gait Detection Monitor for Sled Dogs” by Arianna Mastali (Georgia Institute of Technology). According to her, studies have shown orthopedic injuries to be most common among sled dogs. These like to move very much, but repeatedly exceed their capabilities. To solve this problem, the team has developed a technical solution, a special wearable, with the help of which data on the condition of the animals are generated. “Spatial and Temporal Analytic Pipeline for Evaluation of Potential Guide Dogs Using Location and Behavior Data” was the title of the next talk, given by David L. Roberts (North Carolina State University), followed by “Comparing Accelerometry and Computer Vision Sensing Modalities for High-Resolution Canine Tail Wagging Interpretation”, given by Devon Martin (North Carolina State University). More information on the conference via www.aciconf.org/aci2022.

Posted on

Towards Exploring Perceptions of Dogs

The ACI2022 conference continued on the afternoon of December 7, 2022. “Paper Session 2: Recognising Animals & Animal Behaviour” began with a presentation by Anna Zamansky (University of Haifa). The title was “How Can Technology Support Dog Shelters in Behavioral Assessment: an Exploratory Study”. Her next talk was also about dogs: “Do AI Models ‘Like’ Black Dogs? Towards Exploring Perceptions of Dogs with Vision-Language Models”. She went into detail about OpenAI’s CLIP model, among other things. CLIP is a neural network which learns visual concepts from natural language supervision. She raised the question: “How can we use CLIP to investigate adoptability?” Hugo Jair Escalante (INAOE) then gave a presentation on the topic “Dog emotion recognition from images in the wild: DEBIw dataset and first results”. Emotion recognition using face recognition is still in its infancy with respect to animals, but impressive progress is already being made. The last presentation in the afternoon before the coffee break was “Detecting Canine Mastication: A Wearable Approach” by Charles Ramey (Georgia Institute of Technology). He raised the question: “Can automatic chewing detection measure how detection canines are coping with stress?”. More information on the conference via www.aciconf.org.