An indoor, stationary, bicycle training device that provides advantages over conventional designs of exercise bicycles is provided. The stationary bicycle may include a tilting/pivoting mechanism to orient the indoor bicycle to simulate descending or climbing. The indoor bicycle may include flexible and resilient frame elements to support the indoor training device to move side-to-side under some riding situations thereby simulating the side-to-side swaying motion of an outdoor bicycle under the same riding situations. The indoor bicycle may include several combinations of frame adjustments to provide configurable dimensions of the indoor bicycle to adjust the frame to properly fit the rider, which may be adjusted based on corresponding dimensions of a user's outdoor bicycle. Still other aspects of the stationary bicycle device may aid in creating an “outdoor” feeling while using the device.

Controlling an exercise machine includes receiving a stream of measurements of extension of a component of an exercise machine. It further includes characterizing the stream of measurements including detecting at least one extremum having at least one extremum parameter. It further includes matching the extremum parameter with a previously determined signature associated with a user. It further includes changing an output of the exercise machine based at least in part on the match.

Controlling an exercise machine includes receiving a stream of measurements of extension of a component of an exercise machine. It further includes detecting a first phase of a repeated motion. It further includes detecting a transition to a second phase of the repeated motion. A time constraint is applied to the detection of the transition to the second phase of the repeated motion. It further includes controlling a resistance associated with the second phase of the repeated motion.

Methods, systems, and computer-readable mediums for generating, by an artificial intelligence engine, an alignment plan capable of enabling, during a treatment session, the aligning of a user's body. The method comprises generating machine learning models trained to identify alignment plans. The method also comprises receiving treatment data that comprises a first position of the user's body, aspects of a treatment plan, and one or more attributes of the user. The method further comprises generating the alignment plan by using the machine learning models. The generating is based on at least the aspects of the treatment plan and at least one of the one or more attributes of the user. The alignment plan may comprise a target position of the user's body and one or more elements for adjusting the user's body from the first position to the target position. The method also comprises transmitting the plan to a computing device.

A method for displaying archived exercise classes comprising displaying information about archived exercise classes that can be accessed by a first user via a computer network on a display screen at a first location, wherein the first user can select among a plurality of archived classes, outputting digital video and audio content comprising the selected archived class, detecting a performance parameter for the first user at a particular point in the selected class, displaying the performance parameter on the display screen, and displaying performance parameters from a second user at a second location on the display screen such that at least one of the performance parameters from the first user and at least one of the performance parameters from the second user at the same point in the class are presented for comparison.

A method includes providing a first video file to a plurality of exercise machines, the first video file including content associated with an exercise class. The method also includes receiving user data from the plurality of exercise machines, the user data including respective settings associated with a common performance metric. In such a method, the respective settings are used on the plurality of exercise machines during playback of a particular part of the first video file. The method also includes identifying a timestamp associated with the particular part of the first video file, and generating an executable control corresponding to the performance metric. The method further includes generating a second video file comprising the content and the executable control. In such methods, playback of the second video file causes display of the executable control at a part of the second video file corresponding to the timestamp.

A method of compiling workout information in a workout session includes, at a workout server, receiving, via a network, first login information from a first exercise device; receiving, via the network, first workout information from the first exercise device; recording the first workout information in an active workout session; receiving, via the network, second login information from a second exercise device; receiving, via the network, second workout information from the second exercise device; and recording the second workout information in the active workout session.

Motion sickness is reduced for users of mechanical systems used to practice sports that comprise an immersive virtual reality device. In particular, the use of prediction and interpolation algorithms enable fluid movements to be displayed within the virtual environment.

Methods and systems are described for real-time instruction and coaching using a virtual assistant for interaction with a user. Users may receive feedback inferences provided generally in real-time after collection of video samples from the user device. Neural network architectures and layers may be used to determine motion patterns and temporal aspects of the video samples, as well as detect activities of the foreground user despite background noise. The methods and systems may have various capabilities, including but not limited to live feedback on performed exercise activities, exercise scoring, calorie estimation, and repetition counting.

The present invention relates to a system (S) for monitoring the weight adjustment of a dumbbell, which can be used by a user to perform a gymnastic exercise, comprising: a dumbbell (2), provided with a handle (21), having an axial development along an axis (R), able to rotate around said axis (R), clockwise and counterclockwise, and provided with a first end (211) and a second end (212), a first plurality of weights (P1), wherein each weight can be individually coupled to said first end (211), when said handle (21) rotates in one direction, and decoupled from said first end (211), when said handle (21) rotates in the opposite direction, a second plurality of weights (P2), wherein each weight can be individually coupled to said second end (212), when said handle (21) rotates in one direction, and decoupled from said second end (212), when said handle (21) rotates in the opposite direction, wherein said system (S) is characterised in that it comprises an element provided with a plurality of reflective portions (321.sub.a, b, . . . , n), in that it comprises at least one detection device (5) able to detect data coming from at least one reflective portion (321.sub.k) and able to send said data, and in that it comprises a logic control unit (U) able to receive said data sent by said at least one detection device (5) and to associate to said data a weight of said first (P1) and second (P2) plurality of weights. The present invention also relates to the method for operating such system.