AnswersPublished17 sources
How SeePhys Hub Combines 360-Degree Vision, Touch and Full-Body Motion
Launched by PHANES AI and Tinno Mobile on October 14, 2026, SeePhys Hub combines 360 degree first person vision with wrist, body motion and optional touch data to create richer robot training demonstrations. Three waist and leg trackers combine IMU signals and UWB ranging with head and wrist vision to estimate full...
Published byEdited with GPT-6 LunaImages generated with GPT Image 2
Research answer
Key takeaways
- Launched by PHANES AI and Tinno Mobile on October 14, 2026, SeePhys Hub combines 360 degree first person vision with wrist, body motion and optional touch data to create richer robot training demonstrations.
- Three waist and leg trackers combine IMU signals and UWB ranging with head and wrist vision to estimate full body pose; the headband is reported to weigh up to 220 g, with synchronization error up to 0.5 ms among its...
AI Prompt
openai.comCreate a landscape editorial hero image for this Studio Global article: What is PHANES AI’s SeePhys Hub, launched jointly with Tinno Mobile on October 14, 2026, and how do the partners’ respective roles, its four. Article summary: SeePhys Hub is a wearable, whole-body multimodal data-collection system jointly launched by PHANES AI and Tinno Mobile on October 14, 2026, to turn human demonstrations into robot-training data combining first-person vis. Topic tags: general, general web, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fa
SeePhys Hub is a wearable system for recording human activity as multimodal data for robot learning. Launched jointly by PHANES AI and Tinno Mobile on October 14, 2026, it combines first-person video with hand and body-motion information—and can add tactile glove data when a task calls for it.18
47 The idea is to fill in details a head-mounted video alone may miss, such as body posture and the pressure of a hand on an object.
4
18
What the system captures
The kit includes a headband with four cameras, two wrist bands, three motion trackers worn at the waist and lower legs, and optional tactile gloves. The headband is designed to provide 360-degree first-person coverage, while wrist cameras add close-up views of hand activity. The waist and lower-leg trackers contribute body-motion signals, and the gloves capture hand contact pressure.18
47
PHANES AI handles product design and algorithms; Tinno Mobile contributes experience in cost control, supply chains and quality assurance, according to launch coverage.18
Why add body and touch data to first-person video?
Egocentric video can show what a person sees and does with their hands, but it may not fully describe how their body moved to complete a task. Walking, bending or turning can matter to the demonstration even when a forward-facing camera does not clearly show the person’s posture or lower-body movement. Video also cannot directly record the changing pressure of a hand’s contact with an object.4
18
SeePhys Hub is intended to combine those complementary signals in one recording. Its wrist views can add detail around manipulation, while the body trackers support whole-body pose estimation. The optional gloves add tactile information for tasks where contact pressure matters. Reports describe the system’s output as including first-person vision, hand pose and touch, and whole-body pose and force; they do not establish that the kit directly measures force at every point on the body.6
9
How sparse-node pose estimation works
Instead of attaching a tracker to every body segment, SeePhys Hub uses three body nodes at the waist and lower legs. The system combines their inertial measurement unit (IMU) motion signals and ultra-wideband (UWB) distance measurements between nodes with head and wrist vision to estimate full-body pose.18
That is an estimate produced by sensor fusion, not a direct measurement of every joint. The available launch reporting does not provide independent pose-accuracy results, so the reported capabilities should be understood as product claims rather than validated performance benchmarks.18
Reported weight, recording and synchronization specifications
Launch coverage reports that the headband weighs no more than 220 grams and each wrist band no more than 75 grams. The headband and wrist bands are reported to support up to 24 hours of continuous collection. The stated synchronization error of no more than 0.5 milliseconds applies to the headband’s cameras, IMU and UWB; it should not be assumed to describe timing across every component in the full kit.47
Why tactile gloves are optional
Gloves can add pressure data, but may interfere with some activities, such as washing dishes; reporting also notes that tactile sensors can introduce noise. The gloves and wrist bands can be used separately, letting a collection setup include gloves when tactile supervision is useful and leave them off when they could impede the task.12
20
Some coverage says the team is exploring how to estimate touch from vision. The available reporting does not establish the method’s accuracy or show that it can replace tactile sensors, so it is best understood as an area of research rather than a demonstrated system capability.18
From recordings to robot-learning data
SeePhys Hub’s companion platform is reported to parse and clean recordings, split them into action segments and add semantic labels. It also connects the data to workflows for scene reconstruction, whole-body pose estimation, hand-action tracking and motion retargeting for robot learning.9
47
PHANES AI positions the system between lightweight internet video and more demanding motion-capture or real-robot data collection: an attempt to retain the flexibility of recording people in everyday settings while capturing additional body and contact information.12 Whether that trade-off delivers accurate, scalable training data—or improves robot performance—remains to be established by independent evaluation.
Studio Global AI
Continue your research
This page includes a source-backed answer you can continue inside Studio Global.
People also ask
What is the short answer to "How SeePhys Hub Combines 360-Degree Vision, Touch and Full-Body Motion"?
Launched by PHANES AI and Tinno Mobile on October 14, 2026, SeePhys Hub combines 360 degree first person vision with wrist, body motion and optional touch data to create richer robot training demonstrations.
What are the key points to validate first?
Launched by PHANES AI and Tinno Mobile on October 14, 2026, SeePhys Hub combines 360 degree first person vision with wrist, body motion and optional touch data to create richer robot training demonstrations. Three waist and leg trackers combine IMU signals and UWB ranging with head and wrist vision to estimate full body pose; the headband is reported to weigh up to 220 g, with synchronization error up to 0.5 ms among its...