SignerX/SignVerse-2M
SignVerse-2M SignVerse-2M: A Two-Million-Clip Pose-Native Universe of 55+ Sign Languages Links: [Paper] | [Data Files] | [Project Page] SignVerse-2M is a large-scale multilingual pose-native dataset for sign language research. The dataset reorganizes publicly available sign language videos into a unified DWPose-based representation and releases the result as approximately 2 million clips from 39,196 videos covering 55+ sign languages. Rather than… See the full description on the dataset page: https://huggingface.co/datasets/SignerX/SignVerse-2M.
101.9k
1#!/usr/bin/env bash2#SBATCH --job-name=dwpose-cpu-gpu-bench3#SBATCH --nodes=14#SBATCH --ntasks=15#SBATCH --cpus-per-task=326#SBATCH --gres=gpu:17#SBATCH --mem=32G8#SBATCH --time=02:00:009#SBATCH --output=/home/sf895/SignVerse-2M-runtime/slurm/logs/dwpose_cpu_gpu_bench_%j.out10#SBATCH --error=/home/sf895/SignVerse-2M-runtime/slurm/logs/dwpose_cpu_gpu_bench_%j.err11 12set -euo pipefail13 14ROOT_DIR="/cache/home/sf895/SignVerse-2M"15CONDA_SH="/home/sf895/miniconda3/etc/profile.d/conda.sh"16CONDA_ENV="signx2"17SCRIPT="$ROOT_DIR/debug/benchmark_dwpose_cpu_vs_gpu.py"18VIDEO_ID="${1:?video id required}"19FPS="${FPS:-24}"20MAX_FRAMES="${MAX_FRAMES:-240}"21CPU_THREADS="${CPU_THREADS:-1,2,4,8,16,32}"22RAW_VIDEO_DIR="${RAW_VIDEO_DIR:-/home/sf895/SignVerse-2M-runtime/raw_video}"23 24source "$CONDA_SH"25 26printf 'video_id=%s\n' "$VIDEO_ID"27printf 'hostname=%s\n' "$(hostname)"28printf 'cuda_visible_devices=%s\n' "${CUDA_VISIBLE_DEVICES:-unset}"29printf 'fps=%s\n' "$FPS"30printf 'max_frames=%s\n' "$MAX_FRAMES"31printf 'cpu_threads=%s\n' "$CPU_THREADS"32 33conda run -n "$CONDA_ENV" python "$SCRIPT" \34 --video-id "$VIDEO_ID" \35 --raw-video-dir "$RAW_VIDEO_DIR" \36 --fps "$FPS" \37 --max-frames "$MAX_FRAMES" \38 --cpu-threads "$CPU_THREADS"39