Core i7 7700hq

For the virtual experiment, the Unity 2019.1.10f1 version was used as a design tool, with C# as a programming language, running on a computer with Windows 10 Home, having a 64-bit operating system, an Intel Core i7 -7700HQ, 2.8 GHz, 16 GB RAM, and an NVIDIA GeForce GTX 1070 GDDR5 graphics card. A single-board computer was used, the Raspberry Pi (Raspberry Pi Foundation, Cambridge, England, UK), model B 2018 [23 ], controlling a Raspberry Pi camera (Version 2.1 [24 ] with a capability of 120 Hz,) for the end-to-end direct latency tests. Eye-tracking data was collected in a virtual environment using the HTC Vive Pro Eye [25 ] with built-in Tobii eye tracker (Core SW 2.16.4.67) with an accuracy estimation of 0.5°–1.1° and a sampling frequency of 120 Hz (HTC Corporation, Taoyuan, Taiwan). Tobii Pro SDK v1.7.1.1081 [26 ] (Tobii Technology, Stockholm, Sweden) and Vive SRanipal SDK v1.1.0.1 [27 ] (HTC Corporation, Taoyuan, Taiwan) are used to access non-filtered and filtered eye-tracking data, respectively. The embedded HMD’s calibration system is used to calibrate eye-tracking data for each participant.
The HTC Vive headset contains two active-matrix organic light-emitting diode (AMOLED) screens, with a resolution of 2.880 × 1.600 pixels in total with a refresh rate of 90 Hz and a field of view of 110°.

The processor used in this study was an Intel Core i7-7700HQ, with a 16 GB RAM and an 8 GB NVIDIA GTX 1070 GPU. We trained the network on Ubuntu 16.04, and Python 3.6 was used in the training and testing of the MS-ADS network model.
The original Mask Scoring R-CNN model pre-trained on the COCO dataset (Lin et al., 2014 ) was used to initialize the MS-ADS to accelerate the training process. The manually annotated apple images were then utilized for training and testing the MS-ADS network. The iteration number was set to 24 epochs. The initial learning rate was set to 0.02 and later decreased by ten times at the 16th and 22nd epochs, respectively. The momentum and weight decay were set to 0.9 and 1 × 10⁻⁴, respectively. The total training time lasted for 3 h and 6 min.
To test the performance of the proposed MS-ADS method on the detection and instance segmentation of apples, precision, recall, F1 score, mean average precision of the detection bounding box (bbox_mAP), mean average precision of the segmentation mask (mask_mAP) and average run time were used to evaluate the method.

An HTC Vive HMD with two lighthouse-stations for motion tracking was used with two HTC Vive’s wands with 6 degrees of freedom (DoF) to facilitate navigation and interactions within the environment (Kourtesis et al., 2019 ). The VR area where the participants were immersed and interacted with the virtual environments was 4.4 m². Additionally, the HMD was connected to a laptop with an Intel Core i7 7700HQ processor at 2.80 GHz, 16 GB RAM, a 4095 MB NVIDIA GeForce GTX 1070 graphics card, a 931 GB TOSHIBA MQ01ABD100 (SATA) hard disk, and Realtek High Definition Audio.