Pith. sign in

REVIEW 4 major objections 6 minor 64 references

PISEP^2: Pseudo Image Sequence Evolution based 3D Pose Prediction

T0 review · 4 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read PISEP2 predicts all future 3D poses in one step from joint coordinates, with lower error than recursive models on two benchmarks.

desk verdict Competent incremental application of video-prediction blocks to skeleton sequences with a sensible non-recursive decoder, but the 'new problem' and 'state-of-the-art' claims outrun a two-baseline comparison on a custom filtered dataset. read the letter →

arxiv 1909.01818 v1 pith:MSXV54BK submitted 2019-09-04 cs.CV

classification cs.CV
keywords 3Dposepredictionskeletonsequencepseudoimagerepresentationnon-recursivedecodingconvolutionalneuralnetworkhumanmotionerroraccumulation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper introduces a new formulation of 3D human pose prediction: given a window of previous joint coordinates, predict future joint coordinates, which can be captured by cheap depth sensors. To make skeleton data usable by video prediction models, it converts each pose into an 18x3 pseudo-image by ordering joints by body parts, then treats the input as an image sequence whose evolution must be inferred. The central architectural claim is that all future poses can be predicted in one step with an encoder-dynamics-decoder network whose decoders are decoupled, avoiding the error accumulation of recursive sequence-to-sequence models. On the G3D and FNTU (a filtered version of NTU RGB+D) benchmarks, the method reports MSE of 0.1199 and 0.1210 and MAE of 1.1101 and 1.1651, better than the reproduced PredCNN and S-TE baselines. A sympathetic reader would take the paper to establish that joint-coordinate pose prediction is tractable and that non-recursive decoding is the main source of the gain.

What carries the argument

The central object is the pseudo image sequence: each 3D pose becomes an 18x3 matrix of joint coordinates arranged by five body parts, so a 10-frame history is a 10-image sequence. The carrying mechanism is the Encoder-Dynamics-Decoder (EDD) network: shared-weight residual multiplicative blocks encode spatial information, cascade multiplicative units model adjacent-frame dynamics hierarchically so layer m/2 sees the global temporal evolution, and decoupled decoders reconstruct each future pose in one step without feeding predictions back. The L1 loss is part of the mechanism because it directly penalizes small pose differences that L2 would shrink.

What would settle it

Train PISEP2 with a recursive decoder that feeds each predicted pose back as input for the next future frame, keeping the same pseudo-image representation, datasets, and splits. If its MSE on the last predicted frames is not worse than the one-step decoder's, the claim that one-step prediction avoids error accumulation is not supported by the architecture comparison.

Watch

Extended reading notes

Core claim

In the paper's own terms, the discovery is that 3D pose prediction with joint coordinate sequences can be modeled as image sequence evolution: each frame's 18 joints become an 18x3 one-channel pseudo-image whose row order (left arm, right arm, trunk, left leg, right leg) preserves local body structure, and the future is produced by an Encoder-Dynamics-Decoder network. The encoder uses stacked residual multiplicative blocks for spatial structure, a cascade multiplicative unit hierarchy captures temporal evolution at multiple scales, and decoupled decoders generate all future frames at once rather than recursively. The paper argues this removes chain structure, reducing computation and preventing errors from compounding, and reports state-of-the-art results on G3D and FNTU with L1 loss. It also reports that the non-recursive architecture transfers to unseen actions better than the recursive baseline.

Load-bearing premise

The load-bearing premise is that arranging joints into the five-part pseudo-image order is what lets convolutional layers exploit local body correlations; the paper's disorder experiments show this ordering improves MSE by only 0.0033 to 0.0064, so if the ordering matters little the claimed advantage rests almost entirely on the non-recursive decoder.

Editorial extensions

If this is right

  • Because only joint coordinates are required, the method can run on data from low-cost depth cameras rather than specialized motion-capture suits.
  • Predicting all future frames in one pass removes the chain structure, and the frame-wise plots show the accuracy gap over the recursive baseline grows at later timesteps.
  • Modeling spatial and temporal information separately lets the network learn general motion dynamics that transfer to unseen action categories, as the pre-trained-on-FNTU, tested-on-G3D experiment shows.
  • L1 loss outperforms L2 on this task because small pose differences are better reflected by absolute error.
  • The architecture achieves the reported benchmark numbers: MSE 0.1199 on G3D and 0.1210 on FNTU, with MAE 1.1101 and 1.1651.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The near-tie between the hand-designed joint order and random orders (0.0033 to 0.0064 MSE difference) suggests the pseudo-image ordering itself contributes little; the non-recursive decoder may carry most of the gain, which a reader could test by fixing the decoder and varying only the joint order.
  • Because the 18x3 image is only 54 pixels, a convolution kernel quickly covers the whole body; a larger or graph-structured spatial representation might give the body-part prior more room to act.
  • The one-step decoder is a general sequence-to-sequence modification: recursive predictors in video prediction, traffic flow, or weather nowcasting could be converted to decoupled decoders to avoid error accumulation, at the cost of summarizing all history in a single dynamics representation.
  • Axis-wise error analysis shows z-coordinates dominate the loss, so per-axis normalization or a weighted loss could change reported results; the paper does not examine this.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper introduces PISEP^2, a framework for predicting future 3D poses from windows of 3D joint coordinate sequences. It encodes each pose as an 18x3 pseudo-image by ordering joints according to body parts, treats the input as a 10-frame image sequence, and uses an Encoder-Dynamics-Decoder network (RMB encoder, CMU dynamics, decoupled RMB decoders) to predict all 10 future frames in one forward pass. The model is trained with L1 loss and evaluated on G3D and a custom filtered NTU RGB+D subset (FNTU), with PredCNN and a reproduced S-TE as baselines. The authors report lower MSE and MAE than both baselines, plus generalization and fine-tuning experiments, and claim state-of-the-art performance.

Significance. If the reported results were established against a broad baseline set, the paper would make a modest but useful engineering contribution: a lightweight, non-recursive decoder that avoids error accumulation and a compact image-like skeletal representation. The paper is commendable for releasing its dataset splits, reporting frame-wise/joint-wise/axis-wise analyses, and including generalization experiments. However, the significance is currently limited because the 'state-of-the-art' claim is based on only two self-selected baselines, the proposed ordering's contribution is small in the authors' own ablations, and no uncertainty estimates are provided. The core architecture is a modification of PredCNN rather than a wholly new model class.

major comments (4)
  1. [IV-C, Table I] The claim that PISEP^2 achieves state-of-the-art performance (abstract; Section IV-C, Table I) is not supported by the evidence presented. The comparison includes only PredCNN [12], a video-prediction model adapted to the pseudo-image input, and S-TE [61], a fully connected model reproduced by the authors. Standard 3D human motion forecasting methods that operate on joint coordinates or related representations (e.g., the RNN-based model in [5] and its successors, QuaterNet, and DCT-based predictors) are cited in the related work but never compared. Because FNTU is a self-filtered subset of NTU RGB+D, no external state-of-the-art numbers exist for it, and the G3D split is not a standard benchmark split. The reported numbers may show improvement over the two chosen baselines, but they do not establish superiority over the state of the art.
  2. [IV-A, Table I] No uncertainty information is reported. Section IV-A describes a single random 70/210 train/test split for G3D and a single random split for FNTU, and Table I gives one scalar per metric per method. The margins over S-TE are modest (e.g., MSE 0.1199 vs 0.1407 on G3D; 0.1210 vs 0.1425 on FNTU), and the disorder ablations in Table II are even smaller. Without multiple seeds, cross-validation, or error bars, the reader cannot tell whether the reported improvements are stable or within run-to-run variation. This is load-bearing because the main quantitative claim rests on a small margin over only one reproduced baseline.
  3. [IV-D, Table II] The paper's own evaluation of the skeletal representation weakens the novelty claim. In Section IV-D and Table II, the body-part ordering (left arm, right arm, trunk, left leg, right leg) improves G3D MSE by only 0.0033 to 0.0064 over two random orderings, and the text admits 'the improvement of our skeletal representation is limit.' This means the spatial-ordering contribution is marginal; the observed gains over the baselines could be attributable mostly to the non-recursive decoder, which is a modest modification of PredCNN (removing the chain and decoupling decoders). The authors should either provide a more informative representation ablation (e.g., comparing against other skeletal representations from the literature and more than two disorder runs) or substantially temper the claims about modeling joint correlations.
  4. [IV-B, II-B] The dismissal of mocap-based pose prediction as 'not comparable' (Introduction and Section II-B) is not convincing, because the S-TE baseline [61] explicitly converts mocap frames to Cartesian joint coordinates, which is exactly the input format used here. Reference [5] (Martinez et al.) is also a standard human motion prediction method and should be compared or its exclusion justified. The distinction between 'mocap frames' and 'joint coordinate sequences' is a matter of data source and coordinate representation, not a fundamentally different problem; the claimed new problem formulation therefore needs a more careful literature positioning.
minor comments (6)
  1. [IV-A, Eqs. (3)-(4)] The metrics are described as normalized by the length of the predictive sequence, but Equations (3) and (4) contain no averaging or division; please write the normalization explicitly (e.g., divide by N times the number of predicted frames).
  2. [III-C, Eq. (2)] Equation (2) writes the loss as a generic norm; since Section IV-D compares L1 and L2 losses, please use \|y - ŷ\|_1 for the proposed loss.
  3. [IV-C] In the paragraph after Table I, 'the MSE increases by 0.0208 and 0.0215' should read 'decreases' (or 'is lower by'), because PISEP^2 has smaller MSE than S-TE on both datasets.
  4. [IV-B] The sentence 'there exists no baseline for comparison' contradicts the related-work discussion, which cites motion prediction methods such as [5] and [61]; please rephrase to say that no baseline was compared under this exact input format.
  5. [IV-A] The FNTU filtering criterion 'selecting the relative forward skeleton' is vague; please give the exact selection rule (e.g., joint orientation thresholds or camera-view criteria) so that the dataset is reproducible.
  6. [II-B] There are several typos and inconsistencies in notation, e.g., 'the author first covert the mocap frame' and inconsistent superscript formatting for PISEP^2; a careful proofread is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation is a standard supervised learning pipeline evaluated on external datasets and independent baselines.

full rationale

I walked the paper's claimed derivation chain: the skeletal representation is a hand-designed rearrangement of joint coordinates (Section III-A, Eq. 1), the EDD network is an adapted CNN encoder-dynamics-decoder (Section III-B), and the loss is a plain L1 norm (Section III-C, Eq. 2). None of these steps is defined in terms of the prediction target, and no fitted parameter is later renamed as a prediction. The reported MSE/MAE results (Table I) come from training on G3D and FNTU external datasets and comparing against PredCNN and S-TE baselines; the baselines are not constructed from the paper's own outputs, and the method's superiority over them is not forced by construction. The paper has no load-bearing self-citations: the cited building blocks (RMB from Kalchbrenner et al. [52], CMU from PredCNN [12], and the S-TE baseline from Butepage et al. [61]) are external prior work, and the authors do not invoke any uniqueness theorem or prior result by the same authors to exclude alternatives. The main weaknesses are evaluation-framing issues rather than circularity: the 'state-of-the-art' claim rests on only two baselines, the FNTU split is a self-filtered subset of NTU RGB+D, and no mainstream joint-coordinate motion forecasting models are compared. These concerns affect how convincing the empirical claim is, but they do not make the derivation equivalent to its inputs. Therefore the appropriate circularity score is 0.

Assumptions & free parameters 5 free parameters · 3 assumptions · 0 invented entities

The method relies on standard deep learning building blocks and a hand-designed joint ordering. The main domain assumptions are that the ordering preserves useful correlations and that the filtered FNTU dataset is representative. No new physical entities or fitted physical constants are introduced.

free parameters (5)
  • learning rate = 0.0001
    Adam optimizer learning rate chosen by hand in Section IV-A.
  • encoder depth = 2 RMBs on G3D, 4 RMBs on FNTU
    Number of residual multiplicative blocks in the encoder, chosen to avoid over-fitting as stated in Section IV-A.
  • decoder depth = 3 RMBs on G3D, 6 RMBs on FNTU
    Number of decoder RMB blocks, set per dataset as described in Section IV-A.
  • window size = 20
    Sequence length of 10 input and 10 output frames with overlap 5, defined in Section IV-A.
  • overlap size = 5
    Sliding window overlap chosen in Section IV-A to ensure continuity.
assumptions (3)
  • domain assumption The pseudo-image representation with the specified joint ordering models inter-joint correlations relevant for pose prediction.
    Stated in Section III-A without independent evidence; the disorder ablation in Table II shows only a small effect.
  • domain assumption Filtering NTU RGB+D to create FNTU yields a valid and representative testbed.
    Section IV-A removes mutual actions and non-frontal skeletons, which may remove hard cases and make the task easier.
  • standard math The RMB and CMU modules from [52] and [12] behave as published and transfer to 18x3 pseudo-images.
    The paper imports these modules without formal verification, relying on the cited papers.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PISEP^2: Pseudo Image Sequence Evolution based 3D Pose Prediction." pith.science (2026). https://pith.science/paper/MSXV54BK

@misc{pith2026190901818,
  author       = {Pith},
  title        = {Pith review of: PISEP^2: Pseudo Image Sequence Evolution based 3D Pose Prediction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MSXV54BK}},
  note         = {Machine review of arXiv:1909.01818}
}
read the original abstract

Pose prediction is to predict future poses given a window of previous poses. In this paper, we propose a new problem that predicts poses using 3D joint coordinate sequences. Different from the traditional pose prediction based on Mocap frames, this problem is convenient to use in real applications due to its simple sensors to capture data. We also present a new framework, PISEP^2 (Pseudo Image Sequence Evolution based 3D Pose Prediction), to address this new problem. Specifically, a skeletal representation is proposed by transforming the joint coordinate sequence into an image sequence, which can model the different correlations of different joints. With this image based skeletal representation, we model the pose prediction as the evolution of image sequence. Moreover, a novel inference network is proposed to predict all future poses in one step by decoupling the decoders in a non-recursive manner. Compared with the recursive sequence to sequence model, we can improve the computational efficiency and avoid error accumulation significantly. Extensive experiments are carried out on two benchmark datasets (e.g. G3D and FNTU). The proposed method achieves the state-of-the-art performance on both datasets, which demonstrates the effectiveness of our proposed method.

Figures

Figures reproduced from arXiv: 1909.01818 by the authors.

Figure 1
Figure 1. pose prediction. ours. (1) Mocap based pose prediction [5]–[8]: on one hand, the input and output of these works are mocap frames, and the human pose is represented as a mocap vector parameterized by the exponential map which is easy to predict to a great extent [9]–[11]; on the other hand, the acquisition of mocap data is difficult and expensive, and it needs lots of preprocessing to visualize its performance [5], … view at source ↗
Figure 2
Figure 2. Temporal evolution of previous frames. (a) Commonly spatio-temporal [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 4
Figure 4. The representation of skeletal data. The left part of the figure is [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (10 more)
Figure 3
Figure 3. Figure 3: Our framework. A. Skeletal Representation Different from image data, the skeleton sequence is a set of joints coordinates. Therefore, in this section, as is shown in [PITH_FULL_IMAGE:figures/full_fig_p004_3.png]
Figure 5
Figure 5. Figure 5: The framework of EDD. To avoid over-fitting, we stack 2 RMBs as the encoder and stack 3 RMBs as the decoder on G3D dataset. And we stack 4 RMBs for the encoder and stack 6 RMBs for the decoder on FNTU dataset. We train all models using Adam optimizer, and our learning …
Figure 6
Figure 6. Figure 6: Frame-wise performance of different methods. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Joint-wise performance of different methods. (a) Joint-wise MSE of [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 10
Figure 10. Figure 10: Frame-wise performance of different architectures. [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: Frame-wise performance of different losses. [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]
Figure 12
Figure 12. Figure 12: is the frame-wise performance on unseen data. Before fine-tuning, as shown in the left part of [PITH_FULL_IMAGE:figures/full_fig_p010_12.png]
Figure 13
Figure 13. Figure 13: Similarly, the errors of the joints of the upper limbs [PITH_FULL_IMAGE:figures/full_fig_p010_13.png]
Figure 13
Figure 13. Figure 13: Joint-wise performance of unseen data. (a) Joint-wise MSE of [PITH_FULL_IMAGE:figures/full_fig_p011_13.png]
Figure 15
Figure 15. Figure 15: Visualization of frame-wise performance on unseen data. [PITH_FULL_IMAGE:figures/full_fig_p012_15.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 57 canonical work pages

  1. [61]

    Deep repre- sentation learning for human motion prediction and classification,

    J. Butepage, M. J. Black, D. Kragic, and H. Kjellstrom, “Deep repre- sentation learning for human motion prediction and classification,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 6158–6166

  2. [12]

    Predcnn: Predictive learning with cascade convolutions

    Z. Xu, Y . Wang, M. Long, J. Wang, and M. KLiss, “Predcnn: Predictive learning with cascade convolutions.” in International Joint Conference on Artificial Intelligence , 2018, pp. 2940–2947

  3. [5]

    On human motion prediction using recurrent neural networks,

    J. Martinez, M. J. Black, and J. Romero, “On human motion prediction using recurrent neural networks,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017, pp. 2891–2900

  4. [1]

    Human action recognition and prediction: A survey,

    Y . Kong and Y . Fu, “Human action recognition and prediction: A survey,” arXiv preprint arXiv:1806.11230 , 2018

  5. [2]

    Enhanced computer vision with microsoft kinect sensor: A review,

    J. Han, L. Shao, D. Xu, and J. Shotton, “Enhanced computer vision with microsoft kinect sensor: A review,” IEEE Transactions on Cybernetics , vol. 43, no. 5, pp. 1318–1334, 2013

  6. [3]

    Lifting from the deep: Convolu- tional 3d pose estimation from a single image,

    D. Tome, C. Russell, and L. Agapito, “Lifting from the deep: Convolu- tional 3d pose estimation from a single image,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017, pp. 2500–2509

  7. [4]

    A deep learning framework for character motion synthesis and editing,

    D. Holden, J. Saito, and T. Komura, “A deep learning framework for character motion synthesis and editing,” ACM Transactions on Graphics, vol. 35, no. 4, p. 138, 2016

  8. [6]

    Action- agnostic human pose forecasting,

    H.-k. Chiu, E. Adeli, B. Wang, D.-A. Huang, and J. C. Niebles, “Action- agnostic human pose forecasting,” in 2019 IEEE Winter Conference on Applications of Computer Vision . IEEE, 2019, pp. 1423–1432

Show all 64 references
  1. [7]

    Recurrent network models for human dynamics,

    K. Fragkiadaki, S. Levine, P. Felsen, and J. Malik, “Recurrent network models for human dynamics,” in Proceedings of the IEEE International Conference on Computer Vision , 2015, pp. 4346–4354

  2. [8]

    Structural-rnn: Deep learning on spatio-temporal graphs,

    A. Jain, A. R. Zamir, S. Savarese, and A. Saxena, “Structural-rnn: Deep learning on spatio-temporal graphs,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2016, pp. 5308–5317

  3. [9]

    Practical parameterization of rotations using the expo- nential map,

    F. S. Grassia, “Practical parameterization of rotations using the expo- nential map,” Journal of Graphics Tools, vol. 3, no. 3, pp. 29–48, 1998

  4. [10]

    Modeling human motion using binary latent variables,

    G. W. Taylor, G. E. Hinton, and S. T. Roweis, “Modeling human motion using binary latent variables,” in Advances in Neural Information Processing Systems, 2007, pp. 1345–1352

  5. [11]

    Human3. 6m: Large scale datasets and predictive methods for 3d human sensing in natural environments,

    C. Ionescu, D. Papava, V . Olaru, and C. Sminchisescu, “Human3. 6m: Large scale datasets and predictive methods for 3d human sensing in natural environments,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, no. 7, pp. 1325–1339, 2013

  6. [13]

    Flexible spatio-temporal networks for video prediction,

    C. Lu, M. Hirsch, and B. Scholkopf, “Flexible spatio-temporal networks for video prediction,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017, pp. 6523–6531

  7. [14]

    Learning to decompose and disentangle representations for video prediction,

    J.-T. Hsieh, B. Liu, D.-A. Huang, L. F. Fei-Fei, and J. C. Niebles, “Learning to decompose and disentangle representations for video prediction,” in Advances in Neural Information Processing Systems , 2018, pp. 517–526

  8. [15]

    Stochastic variational video prediction,

    M. Babaeizadeh, C. Finn, D. Erhan, R. H. Campbell, and S. Levine, “Stochastic variational video prediction,” in International Conference on Learning Representations , 2017

  9. [16]

    Prediction and tracking of moving objects in image sequences,

    A. G. Bors and I. Pitas, “Prediction and tracking of moving objects in image sequences,” IEEE Transactions on Image Processing , vol. 9, no. 8, pp. 1441–1445, 2000

  10. [17]

    Block-based spatial prediction and transforms based on 2d markov processes for image and video compression,

    F. Kamisli, “Block-based spatial prediction and transforms based on 2d markov processes for image and video compression,” IEEE Transactions on Image Processing , vol. 24, no. 4, pp. 1247–1260, 2015

  11. [18]

    A new representation of skeleton sequences for 3d action recognition,

    Q. Ke, M. Bennamoun, S. An, F. Sohel, and F. Boussaid, “A new representation of skeleton sequences for 3d action recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 3288–3297

  12. [19]

    Co-occurrence feature learning from skeleton data for action recognition and detection with hierarchical aggregation,

    C. Li, Q. Zhong, D. Xie, and S. Pu, “Co-occurrence feature learning from skeleton data for action recognition and detection with hierarchical aggregation,” inInternational Joint Conference on Artificial Intelligence, 2018, pp. 786–792

  13. [20]

    Skeleton based action recognition with convolutional neural network,

    Y . Du, Y . Fu, and L. Wang, “Skeleton based action recognition with convolutional neural network,” in 2015 3rd IAPR Asian Conference on Pattern Recognition. IEEE, 2015, pp. 579–583

  14. [21]

    Skeleton-based action recognition with convolutional neural networks,

    C. Li, Q. Zhong, D. Xie, and S. Pu, “Skeleton-based action recognition with convolutional neural networks,” in 2017 IEEE International Con- ference on Multimedia & Expo Workshops . IEEE, 2017, pp. 597–600

  15. [22]

    Joint distance maps based action recognition with convolutional neural networks,

    C. Li, Y . Hou, P. Wang, and W. Li, “Joint distance maps based action recognition with convolutional neural networks,” IEEE Signal Processing Letters, vol. 24, no. 5, pp. 624–628, 2017

  16. [23]

    Representation learning of temporal dynamics for skeleton-based action recognition,

    Y . Du, Y . Fu, and L. Wang, “Representation learning of temporal dynamics for skeleton-based action recognition,” IEEE Transactions on Image Processing, vol. 25, no. 7, pp. 3010–3022, 2016. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 13

  17. [24]

    Learning clip representations for skeleton-based 3d action recognition,

    Q. Ke, M. Bennamoun, S. An, F. Sohel, and F. Boussaid, “Learning clip representations for skeleton-based 3d action recognition,” IEEE Transactions on Image Processing, vol. 27, no. 6, pp. 2842–2855, 2018

  18. [25]

    View-invariant human action recognition based on a 3d bio-constrained skeleton model,

    Q. Nie, J. Wang, X. Wang, and Y . Liu, “View-invariant human action recognition based on a 3d bio-constrained skeleton model,” IEEE Transactions on Image Processing , 2019

  19. [26]

    Timeception for complex action recognition,

    N. Hussein, E. Gavves, and A. W. Smeulders, “Timeception for complex action recognition,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2019, pp. 254–263

  20. [27]

    Two-stream convolutional networks for action recognition in videos,

    K. Simonyan and A. Zisserman, “Two-stream convolutional networks for action recognition in videos,” in Advances in Neural Information Processing Systems, 2014, pp. 568–576

  21. [28]

    Rethinking spatiotem- poral feature learning: Speed-accuracy trade-offs in video classification,

    S. Xie, C. Sun, J. Huang, Z. Tu, and K. Murphy, “Rethinking spatiotem- poral feature learning: Speed-accuracy trade-offs in video classification,” in Proceedings of the European Conference on Computer Vision , 2018, pp. 305–321

  22. [29]

    Appearance-and-relation net- works for video classification,

    L. Wang, W. Li, W. Li, and L. Van Gool, “Appearance-and-relation net- works for video classification,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2018, pp. 1430–1439

  23. [30]

    A closer look at spatiotemporal convolutions for action recognition,

    D. Tran, H. Wang, L. Torresani, J. Ray, Y . LeCun, and M. Paluri, “A closer look at spatiotemporal convolutions for action recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 6450–6459

  24. [31]

    Hierarchical recurrent neural network for skeleton based action recognition,

    Y . Du, W. Wang, and L. Wang, “Hierarchical recurrent neural network for skeleton based action recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2015, pp. 1110–1118

  25. [32]

    Bio- inspired dynamic 3d discriminative skeletal features for human action recognition,

    R. Chaudhry, F. Ofli, G. Kurillo, R. Bajcsy, and R. Vidal, “Bio- inspired dynamic 3d discriminative skeletal features for human action recognition,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2013, pp. 471–478

  26. [33]

    Few-shot human motion prediction via meta-learning,

    L.-Y . Gui, Y .-X. Wang, D. Ramanan, and J. M. Moura, “Few-shot human motion prediction via meta-learning,” in Proceedings of the European Conference on Computer Vision , 2018, pp. 432–450

  27. [34]

    Predrnn: Recurrent neural networks for predictive learning using spatiotemporal lstms,

    Y . Wang, M. Long, J. Wang, Z. Gao, and S. Y . Philip, “Predrnn: Recurrent neural networks for predictive learning using spatiotemporal lstms,” in Advances in Neural Information Processing Systems , 2017, pp. 879–888

  28. [35]

    G3d: A gaming action dataset and real time action recognition evaluation framework,

    V . Bloom, D. Makris, and V . Argyriou, “G3d: A gaming action dataset and real time action recognition evaluation framework,” in 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recog- nition Workshops. IEEE, 2012, pp. 7–12

  29. [36]

    Ntu rgb+ d: A large scale dataset for 3d human activity analysis,

    A. Shahroudy, J. Liu, T.-T. Ng, and G. Wang, “Ntu rgb+ d: A large scale dataset for 3d human activity analysis,” in Proceedings of the IEEE conference on Computer Vision and Pattern Recognition , 2016, pp. 1010–1019

  30. [37]

    Deep predictive coding networks for video prediction and unsupervised learning,

    W. Lotter, G. Kreiman, and D. Cox, “Deep predictive coding networks for video prediction and unsupervised learning,” in International Con- ference on Learning Representations , 2017

  31. [38]

    One-step time-dependent future video frame prediction with a convo- lutional encoder-decoder neural network,

    V . Vukoti´c, S.-L. Pintea, C. Raymond, G. Gravier, and J. C. Van Gemert, “One-step time-dependent future video frame prediction with a convo- lutional encoder-decoder neural network,” in International Conference on Image Analysis and Processing . Springer, 2017, pp. 140–151

  32. [39]

    Frequency domain transformer networks for video prediction,

    H. Farazi and S. Behnke, “Frequency domain transformer networks for video prediction,” arXiv preprint arXiv:1903.00271 , 2019

  33. [40]

    Patch to the future: Unsupervised visual prediction,

    J. Walker, A. Gupta, and M. Hebert, “Patch to the future: Unsupervised visual prediction,” in Proceedings of the IEEE conference on Computer Vision and Pattern Recognition , 2014, pp. 3302–3309

  34. [41]

    Deep multi-scale video prediction beyond mean square error,

    M. Mathieu, C. Couprie, and Y . LeCun, “Deep multi-scale video prediction beyond mean square error,” arXiv preprint arXiv:1511.05440, 2015

  35. [42]

    Action-conditional video prediction using deep networks in atari games,

    J. Oh, X. Guo, H. Lee, R. L. Lewis, and S. Singh, “Action-conditional video prediction using deep networks in atari games,” in Advances in Neural Information Processing Systems , 2015, pp. 2863–2871

  36. [43]

    Deep spatio-temporal residual networks for citywide crowd flows prediction,

    J. Zhang, Y . Zheng, and D. Qi, “Deep spatio-temporal residual networks for citywide crowd flows prediction,” in Thirty-First AAAI Conference on Artificial Intelligence , 2017

  37. [44]

    Dnn-based prediction model for spatio-temporal data,

    J. Zhang, Y . Zheng, D. Qi, R. Li, and X. Yi, “Dnn-based prediction model for spatio-temporal data,” in Proceedings of the 24th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. ACM, 2016, p. 92

  38. [45]

    Convolutional lstm network: A machine learning approach for precipitation nowcasting,

    S. Xingjian, Z. Chen, H. Wang, D.-Y . Yeung, W.-K. Wong, and W.- c. Woo, “Convolutional lstm network: A machine learning approach for precipitation nowcasting,” inAdvances in Neural Information Processing Systems, 2015, pp. 802–810

  39. [46]

    Unsupervised learning for physical interaction through video prediction,

    C. Finn, I. Goodfellow, and S. Levine, “Unsupervised learning for physical interaction through video prediction,” in Advances in Neural Information Processing Systems , 2016, pp. 64–72

  40. [47]

    Predrnn++: Towards a resolution of the deep-in-time dilemma in spatiotemporal predictive learning,

    Y . Wang, Z. Gao, M. Long, J. Wang, and P. S. Yu, “Predrnn++: Towards a resolution of the deep-in-time dilemma in spatiotemporal predictive learning,” in International Conference on Machine Learning , 2018

  41. [48]

    Dual motion gan for future-flow embedded video prediction,

    X. Liang, L. Lee, W. Dai, and E. P. Xing, “Dual motion gan for future-flow embedded video prediction,” in Proceedings of the IEEE International Conference on Computer Vision , 2017, pp. 1744–1752

  42. [49]

    Stochastic adversarial video prediction,

    A. X. Lee, R. Zhang, F. Ebert, P. Abbeel, C. Finn, and S. Levine, “Stochastic adversarial video prediction,” arXiv preprint arXiv:1804.01523, 2018

  43. [50]

    Structure preserving video prediction,

    J. Xu, B. Ni, Z. Li, S. Cheng, and X. Yang, “Structure preserving video prediction,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 1460–1469

  44. [51]

    Deep generative video prediction,

    T. Yu, L. Wang, H. Gu, S. Xiang, and C. Pan, “Deep generative video prediction,” Pattern Recognition Letters, vol. 110, pp. 58–65, 2018

  45. [52]

    Video pixel networks,

    N. Kalchbrenner, A. van den Oord, K. Simonyan, I. Danihelka, O. Vinyals, A. Graves, and K. Kavukcuoglu, “Video pixel networks,” in Proceedings of the 34th International Conference on Machine Learning , vol. 70. JMLR. org, 2017, pp. 1771–1779

  46. [53]

    Decomposing motion and content for natural video sequence prediction,

    R. Villegas, J. Yang, S. Hong, X. Lin, and H. Lee, “Decomposing motion and content for natural video sequence prediction,” in International Conference on Learning Representations , 2017

  47. [54]

    Deep video generation, prediction and completion of human action sequences,

    H. Cai, C. Bai, Y .-W. Tai, and C.-K. Tang, “Deep video generation, prediction and completion of human action sequences,” in Proceedings of the European Conference on Computer Vision , 2018, pp. 366–382

  48. [55]

    The pose knows: Video forecasting by generating pose futures,

    J. Walker, K. Marino, A. Gupta, and M. Hebert, “The pose knows: Video forecasting by generating pose futures,” in Proceedings of the IEEE International Conference on Computer Vision , 2017, pp. 3332– 3341

  49. [56]

    Human pose forecasting via deep markov models,

    S. Toyer, A. Cherian, T. Han, and S. Gould, “Human pose forecasting via deep markov models,” in 2017 International Conference on Digital Image Computing: Techniques and Applications. IEEE, 2017, pp. 1–8

  50. [57]

    Forecasting human dynamics from static images,

    Y .-W. Chao, J. Yang, B. Price, S. Cohen, and J. Deng, “Forecasting human dynamics from static images,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 548– 556

  51. [58]

    Structured prediction of 3d human pose with deep neural networks,

    B. Tekin, I. Katircioglu, M. Salzmann, V . Lepetit, and P. Fua, “Structured prediction of 3d human pose with deep neural networks,” arXiv preprint arXiv:1605.05180, 2016

  52. [59]

    Forecasting human pose and motion with multi- body dynamic model,

    S. Cao and R. Nevatia, “Forecasting human pose and motion with multi- body dynamic model,” in 2015 IEEE Winter Conference on Applications of Computer Vision . IEEE, 2015, pp. 191–198

  53. [60]

    Real-time human motion forecasting using a rgb camera,

    E. Wu and H. Koike, “Real-time human motion forecasting using a rgb camera,” in Proceedings of the 24th ACM Symposium on Virtual Reality Software and Technology. ACM, 2018, p. 118

  54. [62]

    Adversarial geometry-aware human motion prediction,

    L.-Y . Gui, Y .-X. Wang, X. Liang, and J. M. Moura, “Adversarial geometry-aware human motion prediction,” in Proceedings of the Euro- pean Conference on Computer Vision , 2018, pp. 786–803

  55. [63]

    Long-term human motion prediction by modeling motion context and enhancing motion dynamic,

    Y . Tang, L. Ma, W. Liu, and W. Zheng, “Long-term human motion prediction by modeling motion context and enhancing motion dynamic,” arXiv preprint arXiv:1805.02513 , 2018

  56. [64]

    Bio-inspired predictive orientation decom- position of skeleton trajectories for real-time human activity prediction,

    H. Zhang and L. E. Parker, “Bio-inspired predictive orientation decom- position of skeleton trajectories for real-time human activity prediction,” in 2015 IEEE International Conference on Robotics and Automation . IEEE, 2015, pp. 3053–3060

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.