The infogcn2 from stnoah1

infogcn2's Introduction

👋 Hi, I am Hyung-gun Chi (지형근) <🏠|🎓|📄>

I am a PhD student in Computer Engineering at Purdue University, advised by Professor Karthik Ramani. My research interests lie in Computer Vision, focusing on 3D Geometric Deep Learning and Action Analysis and their applications to AR/VR.

📄 Selected Publications

W. Roh, S. Lee, W. Ryoo, G. Oh, J. Lee, S. Hwang, H. Chi, and S. Kim. Functional Hand Type Prior for 3D Hand Pose Estimation and Action Recognition from Egocentric View Monocular Videos. BMVC (Oral), 2023.
H. Chi, K. Lee, N. Agarwal, K. Ramani, and C. Choi. AdamsFormer for Spatial Action Localization in the Future. CVPR, 2023.
Y. Xu, A. Bazarjani, H. Chi, C. Choi, and Y. Fu. Uncovering the Missing Pattern: Unified Framework Towards Trajectory Imputation and Prediction. CVPR, 2023.
H. Chi*, S. Chi*, S. Chan, and Karthik Ramani. Pose Relation Transformer : Refine Occlusions for Human Pose Estimation. ICRA, 2023. [CODE]
H. Chi*, M. H. Ha*, S. Chi, S. Lee, Q. Huang, and Karthik Ramani. InfoGCN : Representation Learning for Human Skeleton-based Action Recognition. CVPR, 2022. [CODE]
H. Chi*, S. Kim*, X. Hu, Q. Huang, and Karthik Ramani. A Large-scale Annotated Mechanical Components Benchmark for Classification and Retrieval Tasks with Deep Neural Networks. ECCV, 2020. [CODE]
S. Kim, H. Chi, and Karthik Ramani. First-Person View Hand Segmentation of Multi-Modal Hand Activity Video Dataset. BMVC, 2020.

💻 Skills

Language : Python, Matlab, C/C++, SQL, JavaScript, HTML, CSS, PHP
Machine/Deep Learning : PyTorch, TensorFlow, Keras
Web Programming : MYSQL/mongoDB, Flask/Node.js
CAD Tool : Creo Parametric, SolidWorks, AutoCAD
ETC : GAZEBO, ROS

infogcn2's People

Contributors

Stargazers

Watchers

infogcn2's Issues

I did not find your PORT github from your link https://github.com/stnoah1/PORT.

Could you provide this PORT code in github like you mentioned in your paper?

IndexError: tuple index out of range

Hi, thanks for your awesome work! Here occurs a problem when I try to train InfoGCN++ on NTU120 dataset.
Here are some details:

Traceback (most recent call last):
  File "main.py", line 464, in <module>
    main()
  File "main.py", line 461, in main
    processor.start()
  File "main.py", line 428, in start
    self.train(epoch, save_model=save_model)
  File "main.py", line 241, in train
    y_hat, x_hat, z_0, z_hat, kl_div = self.model(x)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/infoGCN--/model/sode.py", line 244, in forward
    z_0, z_hat, z_hat_shifted = self.extrapolate(z, self.arange_n_step.to(z.dtype))
  File "/home/infoGCN--/model/sode.py", line 182, in extrapolate
    zs = self.diffeq_solver(z_0, t) # z_i = 2, (b t), c, v
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/infoGCN--/model/sode.py", line 36, in forward
    pred_y = odeint(self.ode_func, first_point, time_steps_to_predict,
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torchdiffeq/_impl/odeint.py", line 77, in odeint
    solution = solver.integrate(t)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torchdiffeq/_impl/solvers.py", line 105, in integrate
    dy, f0 = self._step_func(self.func, t0, dt, t1, y0)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torchdiffeq/_impl/fixed_grid.py", line 10, in _step_func
    f0 = func(t0, y0, perturb=Perturb.NEXT if self.perturb else Perturb.NONE)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torchdiffeq/_impl/misc.py", line 189, in forward
    return self.base_func(t, y)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/infoGCN--/model/sode.py", line 59, in forward
    x = self.conv1(x)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 313, in forward
    return self._conv_forward(input, self.weight, self.bias)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 309, in _conv_forward
    return F.conv1d(input, weight, bias, self.stride,
  File "/usr/local/miniconda3/lib/python3.8/site-packages/apex/amp/wrap.py", line 21, in wrapper
    args[i] = utils.cached_cast(cast_fn, args[i], handle.cache)
  File "/usr/local/miniconda3/lib/python3.8/site-packages/apex/amp/utils.py", line 97, in cached_cast
    if cached_x.grad_fn.next_functions[1][0].variable is not x:
IndexError: tuple index out of range

The hyper parameters are as followed

{'debug': False, 'log_dir': '.', 'model_saved_name': '', 'noise_ratio': 0.5, 'project': 'SODE', 'n_desired': 40000, 'num_point': 25, 'num_person': 2, 'num_class': 120, 'dataset': 'ntu120', 'datacase': 'NTU120_CSet', 'use_vel': False, 'phase': 'train', 'save_score': True, 'seed': 1, 'log_interval': 100, 'save_interval': 1, 'save_epoch': 0, 'eval_interval': 5, 'print_log': True, 'show_topk': [1, 5], 'feeder': 'feeders.feeder_ntu.Feeder', 'num_worker': 8, 'random_rot': True, 'repeat': 1, 'window_size': 64, 'base_channel': 64, 'weights': None, 'ignore_weights': [], 'n_heads': 3, 'depth': 4, 'k': 8, 'z_prior_gain': 3, 'graph': 'graph.ntu_rgb_d.Graph', 'dct': False, 'n_sample': 1, 'SAGC_proj': True, 'backbone': 'transformer', 'num_cls': 120, 'base_lr': 0.1, 'step': [50, 60], 'optimizer': 'SGD', 'nesterov': True, 'batch_size': 64, 'test_batch_size': 64, 'start_epoch': 0, 'num_epoch': 110, 'weight_decay': 0.0003, 'lr_decay_rate': 0.1, 'warm_up_epoch': 0, 'lambda_1': 1.0, 'lambda_2': 0.1, 'lambda_3': 0.1, 'lambda_4': 0.0, 'lambda_5': 0.0, 'half': True, 'amp_opt_level': 1, 'n_step': 3, 'n_min': 3, 'dilation': 1, 'ode_method': 'euler', 'work_dir': '/home/infoGCN--/wandb/run-20240102_065914-eeptaw4a/files'}

Recommend Projects