Computer Vision and Pattern Recognition · arXiv:2611.16793 v1 · submitted 12 Nov 2026 · 63 references
EG360: Efficient and Generalizable 360° Depth Estimation via Adaptive Panorama-Perspective Distillation and Spherical Cross-Level Feature Fusion
The paper says
“Generalized panorama depth estimation has witnessed great advances in recent years. These methods typically achieve robust zero-shot generalization performance using large pretrained models, but their high computational cost makes real-time inference on embedded platforms challenging.”
more of the abstract
In contrast, previous lightweight panorama models have mostly been developed for single-domain evaluations and cannot generalize well due to domain shifts. To reduce the gap between these two types of models, we introduce EG360, an efficient and generalizable panorama depth estimation model. Specifically, building on a lightweight depth model, we design an Adaptive Panorama-Perspective Distillation (APPD) strategy guided by scene geometry, along with a Spherical Cross-Level Feature Fusion (SCFF) module for global semantic perception. As a result, with only 6.593M parameters after reparameterization, EG360 achieves strong zero-shot performance on real-world benchmarks and runs at 35 FPS on a Jetson Orin NX 16GB in MAXN mode, whereas a previous ViT-small model runs at less than 2 FPS. Our code will also be made public for future research.
from the abstract
Pith says
Machine review · deepseek-v4-flash · 13 Nov 2026 · what was read was not recorded · not checked by a person
It can be wrong. Challenge this review or reply to it.
Holds up, with conditions. Solid engineering paper on efficient panorama depth estimation; the distillation weighting is the main novelty, and the experiments are thorough. The main soft spot is the lack of error bars and sensitivity analysis for the key hyperparameter tau, but the central claim is plausible. This line is wrong
5 major objections · 8 minor
How often objections from reviews like this one are right: not yet measured
Not checked: figures, code, what the 63 references say.
The paper says, at §3.3, Eq. (3)
The review did not quote this passage.
Pith says: Objection 1 of 5 · major
The hyperparameter τ in the teacher weighting formula w_P = τ/(r_c+τ), w_C = r_c/(r_c+τ) is set to 0.3 experimentally (line after Eq. (3)). The paper does not report any sensitivity analysis for τ. Table 7 shows that simple averaging (which corresponds to τ → ∞) already yields strong results, and APPD provides a modest improvement on Stanford2D3D (AbsRel 0.101 vs 0.109) but slightly worse on Matterport3D (0.169 vs 0.168) and Metropolis (0.244 vs 0.238).
more of objection 1
Without a sensitivity study, it is unclear whether the chosen τ is robust across datasets or whether the improvement over averaging is statistically significant. The authors should either provide an ablation over τ values (e.g., τ ∈ {0.1, 0.3, 0.5, 1.0}) or justify why τ=0.3 is a principled choice.
The paper says, at §4.1, Implementation Details
The review did not quote this passage.
Pith says: Objection 2 of 5 · major
The paper states that training runs for 100,000 iterations with a batch size of 8 for labeled data and 8 for unlabeled data. However, the ablation in Table 6 varies the unlabeled batch size while keeping the labeled batch size fixed.
more of objection 2
The paper does not specify whether the total number of iterations or the total number of epochs is held constant when varying the unlabeled batch size. If the number of iterations is fixed, then a smaller unlabeled batch size means fewer total unlabeled samples seen during training, which could explain the performance drop. The authors should clarify the training protocol and ensure that the ablation is fair.
The paper says, at §4.3, Table 7
The review did not quote this passage.
Pith says: Objection 3 of 5 · major
The 'Manual' strategy (assign DAP to ZInD and MoGe2 to HoliCity) achieves AbsRel 0.166 on Matterport3D, which is better than APPD's 0.169. The paper does not discuss this result. If a simple dataset-level assignment outperforms the geometry-adaptive weighting on one dataset, the claim that APPD is 'guided by scene geometry' and provides a 'useful accuracy trade-off' needs qualification.
more of objection 3
The authors should explain why APPD underperforms Manual on Matterport3D and whether this indicates that the planar residual heuristic is not always optimal.
The paper says, at §4.3, Table 10
The review did not quote this passage.
Pith says: Objection 4 of 5 · major
The SCFF ablation is performed on Stanford2D3D using labeled data only (no pseudo-labels). However, the full EG360 model uses both APPD and SCFF together. The interaction between the two components is not studied: does SCFF provide additional benefit when APPD is already applied?
more of objection 4
Table 4 shows that adding SCFF to the baseline improves AbsRel from 0.126 to 0.113 on Stanford2D3D, but when APPD is already present, adding SCFF (EG360 vs +APPD) improves from 0.101 to 0.094. The paper should discuss whether the gains from SCFF are additive or whether there is diminishing returns.
The paper says, at §4.2, Table 3
The review did not quote this passage.
Pith says: Objection 5 of 5 · major
The comparison with foundation models (DAP, DepthMaster, PaGeR) is informative but the paper does not discuss the fact that these models are evaluated using their released weights, which may have been trained on different data distributions. The lightweight models (UniFuse, BiFuse++, ACDNet) are retrained on the same labeled data for fairness, but the foundation models are not.
more of objection 5
This asymmetry should be acknowledged as a limitation when interpreting the accuracy gap. Additionally, DepthMaster and PaGeR ran out of memory (OOM) on the Jetson, so their latency is not reported; this is fine, but the paper should note that the comparison is incomplete.
8 minor objections
The paper says, at Abstract
Pith says: minor objection 1 of 8
The abstract states 'our code will also be made public for future research.' It would be helpful to specify the intended license and repository platform.
The paper says, at §3.3, Eq. (2)
Pith says: minor objection 2 of 8
The normalized plane-fitting residual r_c is defined as the RMS distance to the fitted plane divided by the mean radial depth. The notation uses dist(p_{c,i}, Π_c) but does not define dist explicitly. Consider adding a brief definition (e.g., perpendicular distance).
show 6 more
The paper says, at §3.4, Eq. (6)
Pith says: minor objection 3 of 8
The spherical direction bias uses G_{ij} = 1 - ω_i^T ω_j, which is the cosine distance. The paper states that β = softplus(b) with learnable b. It would be clearer to state that β is learned per head, not shared across heads.
The paper says, at §4.1, Comparison settings
Pith says: minor objection 4 of 8
The paper mentions that 'Attention uses native PyTorch Scaled Dot-Product Attention (SDPA) with external xFormers and flashAttn disabled for fairness.' It would be helpful to note whether this applies to all models or only to EG360.
The paper says, at §4.2, Table 3
Pith says: minor objection 5 of 8
The latency for DA360-S is reported as 517.412 ms, which corresponds to ~1.93 FPS. The abstract states 'a previous ViT-small model runs at less than 2 FPS,' which is consistent. However, the paper should explicitly state that DA360-S is the ViT-small model referred to in the abstract.
The paper says, at §4.3, Table 8
Pith says: minor objection 6 of 8
The LSS loss (least-squares scale alignment) yields very high AbsRel on Metropolis (0.466). The paper attributes this to scale recovery issues. It would be useful to add a brief explanation of why LSS fails on outdoor data.
The paper says, at §5, Limitations
Pith says: minor objection 7 of 8
The paper acknowledges limited unlabeled data and temporal incoherence. It would also be relevant to mention that the method relies on two frozen teacher models, which may become outdated as better teachers emerge, and that the distillation pipeline requires caching their predictions offline.
The paper says, at References
Pith says: minor objection 8 of 8
Reference [26] (Metropolis Dataset) is listed as 'Mapillary. Metropolis Dataset, 2024. Accessed: 2025-11-01.' This is a dataset reference; consider providing a more permanent URL or DOI if available.
What would settle it
Run the APPD weighting with tau=0.1 and tau=1.0 on the Matterport3D validation set; if the AbsRel varies by more than 0.01 across this range, the sensitivity is higher than the paper implies, but even then the central claim of competitive accuracy at 35 FPS would not be overturned.
from the skeptic's reading
How the review went
4 readings by deepseek-v4-flash · 13 Nov 2026
- Reader holds up, with conditions
- Skeptic partly agrees
- Editor wrote the verdict above
- Referee major revision: 5 major objections, 8 minor
Not run on this paper: the simulated authors' defence.
Pith did not check this review's objections against the paper.
The review in full
The review in brief, machine-written
Hey,
This paper is a well-engineered system for real-time panorama depth estimation on embedded devices. The main novelty is the adaptive weighting between a panoramic teacher (DAP) and a perspective teacher (MoGe2) based on plane-fitting residuals per cube face. That's a sensible idea: if a face is mostly planar, the panorama teacher's global context is more useful; if it's geometrically complex, the perspective teacher's local detail helps. The spherical direction bias in the cross-level attention is also a nice touch, though it's a small addition.
The experimental setup is thorough: three zero-shot test sets, comparisons with multiple baselines (some retrained for fairness), and ablation studies for each component. The evaluation is done without scale alignment, which is appropriate. The efficiency numbers are impressive: 6.6M params, 23.5 GFLOPs, 35 FPS on a Jetson Orin NX. That's a real achievement.
The soft spots are real but not fatal. First, there are no error bars anywhere. All results are single runs. Given the stochasticity in training (data augmentation, batch sampling), this is a concern. Second, the key hyperparameter tau (teacher weighting) is set to 0.3 with no sensitivity analysis. The ablation in Table 7 shows that simple averaging (which corresponds to tau → ∞) already gives strong results, and APPD provides a modest additional improvement. So the sensitivity concern is real but not load-bearing: even if tau is suboptimal, the model still works well. Third, code is not yet released, which limits reproducibility. The authors promise to release it, but that's a promise.
The paper is honest about its limitations: limited unlabeled data, single-frame only. The citation pattern looks fine.
Who is this for? Researchers working on efficient panorama depth estimation for robotics or VR. It's a solid engineering contribution. The central claim is plausible and supported by the reported numbers. I'd send it to peer review; it deserves a serious referee. The main things to check in review are the sensitivity of tau and the statistical reliability of the results.
Best, [Your name]
The strongest objection
The paper presents a well-engineered system with thorough ablation studies, clear motivation for each design choice, and honest discussion of limitations. The central claim—that EG360 achieves competitive zero-shot panorama depth accuracy at 35 FPS on an embedded device—is supported by the reported numbers and the ablation tables. The reader's weakest_assumption about tau sensitivity is acknowledged but not a load-bearing concern: the paper states tau is set experimentally (Section 3.3), and the ablation in Table 7 shows that even simple averaging (which corresponds to tau → ∞) already yields strong results, with APPD providing a modest additional improvement. The absence of error bars is a reproducibility concern but does not invalidate the central claim. The code release promise addresses the main verification gap. I find no internal inconsistency or hidden assumption that would undermine the core argument.
The reader's verdict is CONDITIONAL with high confidence, citing the lack of error bars, tau sensitivity analysis, and code availability. I agree that these are reasonable concerns for a reproducibility audit, but I do not find them load-bearing for the central claim. The paper's own ablation (Table 7) shows that simple averaging of teacher losses (which corresponds to tau → ∞) already achieves AbsRel 0.168 on Matterport3D, very close to the full APPD result of 0.169. This indicates that the exact value of tau is not critical to the method's success. The absence of error bars is a reporting weakness but does not make the central claim implausible. The code release is promised and would enable verification. The paper's limitations section honestly acknowledges the limited unlabeled data and temporal incoherence. I see no hidden assumption or internal contradiction. Therefore I recommend UNCHANGED: the CONDITIONAL verdict is appropriate, and my analysis does not suggest a need to strengthen or weaken it.
The referee's report
If the reported results hold, EG360 represents a meaningful step toward practical deployment of panorama depth estimation on resource-constrained devices. The paper demonstrates that a carefully designed distillation strategy combining complementary teachers and a lightweight architecture with spherical-aware global context can narrow the gap between efficient models and large foundation models. The thorough ablation studies (Tables 4-10) provide evidence for the contribution of each component. The code release promise supports reproducibility. The work is likely to be of interest to the community working on 360° perception, efficient deep learning, and embedded vision systems.
Pith's reading of the paper (a machine's summary, not the authors')
The paper tries to establish that a lightweight panorama depth estimation model can achieve strong zero-shot generalization across indoor and outdoor scenes while running in real time on an embedded platform. It argues that prior lightweight models generalize poorly and prior accurate models are too slow, and that the gap can be closed by combining two ideas: adaptively weighting supervision from a panorama teacher and a perspective teacher based on scene planarity, and fusing cross-level features with a spherical direction bias. The paper reports that the resulting model, EG360, with 6.6M parameters, achieves an average AbsRel of 0.168 across Matterport3D, Stanford2D3D, and Metropolis and runs at 35 FPS on a Jetson Orin NX, whereas a ViT-small model runs at less than 2 FPS. A sympathetic reader would care because it suggests that real-time panoramic depth estimation on mobile robots or VR headsets is feasible without sacrificing generalization.
What carries the argument. The key machinery is the normalized plane-fitting residual r_c for each cubemap face, computed from the perspective teacher's 3D points, which quantifies how planar the face is. This scalar drives the adaptive teacher weighting: w_P = tau/(r_c+tau) for the panoramic teacher and w_C = r_c/(r_c+tau) for the perspective teacher. It is the load-bearing mechanism that decides, per face, which teacher's inverse-depth prediction is more reliable, under the assumption that planar views benefit more from panoramic context. The second mechanism is spherical direction bias in cross-level attention, where the attention logit subtracts $\beta$ * G_ij with G_ij = 1 - omega_i^T omega_j, encoding spherical proximity and handling the ERP left-right seam naturally.
Core claim. EG360 demonstrates that a lightweight panorama depth model (6.593M parameters, 23.492 GFLOPs) can achieve competitive zero-shot accuracy on real-world benchmarks while running at 35 FPS on an embedded Jetson Orin NX, narrowing the gap between expensive foundation models and fast but domain-limited lightweight models. The key enablers are: (1) Adaptive Panorama-Perspective Distillation (APPD), which uses the normalized plane-fitting residual of each cubemap face to dynamically weight the contributions of a panoramic teacher (DAP) and a perspective teacher (MoGe2), assigning more weight to the panoramic teacher for planar views; and (2) Spherical Cross-Level Feature Fusion (SCFF), which applies cross-level attention between the two coarsest encoder feature maps with a spherical direction bias that respects the equirectangular projection's left-right continuity and spherical proximity.
The premise it rests on. The distillation hyperparameter tau (set to 0.3) that balances the two teacher weights is chosen experimentally and its sensitivity across different datasets or deployment conditions is not explored; if the optimal tau varies significantly, the claimed robustness may not hold.
If the paper is right (Pith's inferences, not the authors')
Embedded robots and VR headsets can run panoramic depth estimation in real time (35 FPS on Jetson Orin NX) without relying on cloud inference or large models.
The APPD strategy can be applied to other lightweight panorama architectures (UniFuse, BiFuse++, ACDNet) to improve their zero-shot generalization, as shown in Table 9.
The spherical direction bias in SCFF provides a principled way to inject equirectangular geometry into attention mechanisms for any panorama perception task beyond depth.
Training with a simple L1 distillation loss on pseudo labels from complementary teachers outperforms scale-invariant losses (SI, LSS) for zero-shot panorama depth, suggesting that preserving metric scale is important even for unlabeled data.
The normalized plane-fitting residual could be replaced or augmented with a learned planarity score, which might adapt better to non-planar but structured scenes like furniture clusters.
Because the method relies on two frozen teacher models (DAP and MoGe2), its generalization ceiling is tied to those teachers' blind spots; a future lightweight model that jointly trains student and teachers could potentially surpass both.
The spherical direction bias formulation (1 - omega_i^T omega_j) is essentially a cosine-distance kernel on the sphere; it could be interpreted as a learned temperature-adjusted spherical RBF attention, which might generalize to other spherical modalities like 360-degree audio or LIDAR.
Check this review
Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.
How often reviews like this one are right
Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.
How often objections from reviews like this one are right: not yet measured
Reviews by people
This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.
The paper
Abstract
Generalized panorama depth estimation has witnessed great advances in recent years. These methods typically achieve robust zero-shot generalization performance using large pretrained models, but their high computational cost makes real-time inference on embedded platforms challenging. In contrast, previous lightweight panorama models have mostly been developed for single-domain evaluations and cannot generalize well due to domain shifts. To reduce the gap between these two types of models, we introduce EG360, an efficient and generalizable panorama depth estimation model. Specifically, building on a lightweight depth model, we design an Adaptive Panorama-Perspective Distillation (APPD) strategy guided by scene geometry, along with a Spherical Cross-Level Feature Fusion (SCFF) module for global semantic perception. As a result, with only 6.593M parameters after reparameterization, EG360 achieves strong zero-shot performance on real-world benchmarks and runs at 35 FPS on a Jetson Orin NX 16GB in MAXN mode, whereas a previous ViT-small model runs at less than 2 FPS. Our code will also be made public for future research.
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.
abstract compiled · 2026-11-13
The authors' main claim, as the classifier picked it out 0.95
“As a result, with only 6.593M parameters after reparameterization, EG360 achieves strong zero-shot performance on real-world benchmarks and runs at 35 FPS on a Jetson Orin NX 16GB in MAXN mode, whereas a previous ViT-small model runs at less than 2 FPS.”
How the main claim is stated: asserted 1.00
What the abstract itself states, as probabilities: claims first 0.05 · states limitation 0.11 · human participants 0.03 · controlled comparison 0.80 · randomized assignment 0.01 · states quantitative result 0.99 · compares with prior results 0.98 · mentions code or data availability 0.95
The abstract, sentence by sentence
This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.
| Sentence | Role | The authors' own work | States the main conclusion | Strength |
|---|---|---|---|---|
| Generalized panorama depth estimation has witnessed great advances in recent years. | background 1.00 | 0.04 | 0.02 | asserted 1.00 |
| These methods typically achieve robust zero-shot generalization performance using large pretrained models, but their high computational cost makes real-time inference on embedded platforms challenging. | background 0.74 | 0.06 | 0.04 | asserted 1.00 |
| In contrast, previous lightweight panorama models have mostly been developed for single-domain evaluations and cannot generalize well due to domain shifts. | gap 0.72 | 0.05 | 0.03 | asserted 1.00 |
| To reduce the gap between these two types of models, we introduce EG360, an efficient and generalizable panorama depth estimation model. | contribution 1.00 | 0.87 | 0.18 | asserted 0.97 |
| Specifically, building on a lightweight depth model, we design an Adaptive Panorama-Perspective Distillation (APPD) strategy guided by scene geometry, along with a Spherical Cross-Level Feature Fusion (SCFF) module for global semantic perception. | method 0.93 | 0.93 | 0.07 | asserted 0.99 |
| As a result, with only 6.593M parameters after reparameterization, EG360 achieves strong zero-shot performance on real-world benchmarks and runs at 35 FPS on a Jetson Orin NX 16GB in MAXN mode, whereas a previous ViT-small model runs at less than 2 FPS. | result 1.00 | 0.95 | 0.63 | asserted 1.00 |
| Our code will also be made public for future research. | future work 0.89 | 0.33 | 0.03 | asserted 0.93 |
What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.96
What was examined, and what was not
Sources read
- arXiv listing abstract v1
Not examined
- full text
- figures and tables
- references and citing sentences
- relations to other papers
Levels run
- P abstract
Not run
- S sections
- T body sentences
- C citations
- R cross-paper relations
How the labels were made
Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-11-13: 1 request, 5,689 input tokens.
Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON
Figures (6)
Showing 5 of the paper's 6 figures. Figures are the authors', reproduced from the paper.
References (63)
- [1] Hao Ai, Zidong Cao, Yan-Pei Cao, Ying Shan, and Lin Wang. Hrdfuse: Monocular 360 depth estimation by collab- oratively learning holistic-with-regional depth distributions. In2023 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 13273–13282. IEEE, 2023. 2 on Pith
- [2] Iro Armeni, Sasha Sax, Amir R Zamir, and Silvio Savarese. Joint 2d-3d-semantic data for indoor scene understanding. arXiv preprint arXiv:1702.01105, 2017. 1, 3, 5 on PithPith's reviewarXiv
- [3] Shivam Asija, Edward Du, Nam Nguyen, Stefanie Zollmann, and Jonathan Ventura. 3d pano inpainting: Building a vr en- vironment from a single input panorama. In2024 IEEE Con- ference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW), pages 1019–1020. IEEE, 2024. 1 on Pith
- [4] Jimmy Lei Ba, Jamie Ryan Kiros, and Geoffrey E. Hinton. Layer normalization, 2016. 4 on Pith
- [5] Vukasin Bozic, Isidora Slavkovic, Dominik Narnhofer, Nando Metzger, Denis Rozumny, Konrad Schindler, and Nikolai Kalischek. Unified panoramic geometry estima- tion via multi-view foundation models.arXiv preprint arXiv:2605.26368, 2026. 1, 2, 6 on PithPith's reviewarXiv
- [6] Zidong Cao, Jinjing Zhu, Weiming Zhang, Hao Ai, Haotian Bai, Hengshuang Zhao, and Lin Wang. Panda: Towards panoramic depth anything with unlabeled panoramas and m¨obius spatial augmentation. In2025 IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR), pages ... on Pith
- [7] Angel Chang, Angela Dai, Thomas Funkhouser, Maciej Halber, Matthias Niessner, Manolis Savva, Shuran Song, Andy Zeng, and Yinda Zhang. Matterport3d: Learning from rgb-d data in indoor environments.arXiv preprint arXiv:1709.06158, 2017. 1, 5 on PithPith's reviewarXiv
- [8] Sili Chen, Hengkai Guo, Shengnan Zhu, Feihu Zhang, Zi- long Huang, Jiashi Feng, and Bingyi Kang. Video depth anything: Consistent depth estimation for super-long videos. In2025 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 22831–22840. IEEE, 2025. 9 on Pith
- [9] Steve Cruz, Will Hutchcroft, Yuguang Li, Naji Khosravan, Ivaylo Boyadzhiev, and Sing Bing Kang. Zillow indoor dataset: Annotated floor plans with 360° panoramas and 3d room layouts. In2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 2133–2143. ... on Pith
- [10] Tri Dao. FlashAttention-2: Faster attention with better par- allelism and work partitioning. InInternational Conference on Learning Representations (ICLR), 2024. 5 on Pith
- [11] Tri Dao, Daniel Y . Fu, Stefano Ermon, Atri Rudra, and Christopher R´e. FlashAttention: Fast and memory-efficient exact attention with IO-awareness. InAdvances in Neural Information Processing Systems (NeurIPS), 2022. 5 on Pith
- [12] Xiaohan Ding, Xiangyu Zhang, Ningning Ma, Jungong Han, Guiguang Ding, and Jian Sun. Repvgg: Making vgg-style convnets great again. In2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 13728–13737. IEEE, 2021. 2, 6 on Pith
- [13] Alexey Dosovitskiy. An image is worth 16x16 words: Transformers for image recognition at scale.arXiv preprint arXiv:2010.11929, 2020. 1 on PithPith's reviewarXiv
- [14] Krzysztof M Gorski, Eric Hivon, Anthony J Banday, Ben- jamin D Wandelt, Frode K Hansen, Mstvos Reinecke, and Matthia Bartelmann. Healpix: A framework for high- resolution discretization and fast analysis of data distributed on the sphere.The Astrophysical Journal, 622(2):759–771, on Pith
- [15] Xiankang He, Dongyan Guo, Hongji Li, Ruibo Li, Ying Cui, and Chi Zhang. Distill any depth: Distillation creates a stronger monocular depth estimator, 2025. 7 on Pith
- [16] Yuhang He, Wentao Yu, Jie Han, Xing Wei, Xiaopeng Hong, and Yihong Gong. Know your surroundings: Panoramic multi-object tracking by multimodality collaboration. In 2021 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition Workshops (CVPRW), pages 2963–2974. IEEE, 2021. 1 on Pith
- [17] Hualie Jiang, Zhe Sheng, Siyu Zhu, Zilong Dong, and Rui Huang. Unifuse: Unidirectional fusion for 360 panorama depth estimation.IEEE Robotics and Automation Letters, 6 (2):1519–1526, 2021. 1, 2, 6 on Pith
- [18] Hualie Jiang, Ziyang Song, Zhiqiang Lou, Rui Xu, and Minglang Tan. Depth anything in360 ◦: Towards scale in- variance in the wild.arXiv preprint arXiv:2512.22819, 2025. 2, 3, 5, 6, 7, 8 on PithPith's reviewarXiv
- [19] Benjamin Lefaudeux, Francisco Massa, Diana Liskovich, Wenhan Xiong, Vittorio Caggiano, Sean Naren, Min Xu, Jieru Hu, Marta Tintore, Susan Zhang, Patrick Labatut, Daniel Haziza, Luca Wehrstedt, Jeremy Reizenstein, and Grigory Sizov. xformers: A modular and hackable trans- forme... on Pith
- [20] Haodong Li, Wangguangdong Zheng, Jing He, Yuhao Liu, Xin Lin, Xin Yang, Ying-Cong Chen, and Chunchao Guo. DA2: Depth anything in any direction.arXiv preprint arXiv:2509.26618, 2025. 2 on PitharXiv
show 43 more
- [21] Meng Li, Senbo Wang, Weihao Yuan, Weichao Shen, Zhe Sheng, and Zilong Dong.S 2Net: Accurate Panorama Depth Estimation on Spherical Surface.IEEE Robotics and Au- tomation Letters, 8(2):1053–1060, 2023. 2 on Pith
- [22] Haotong Lin, Sili Chen, Junhao Liew, Donny Y Chen, Zhenyu Li, Guang Shi, Jiashi Feng, and Bingyi Kang. Depth anything 3: Recovering the visual space from any views. arXiv preprint arXiv:2511.10647, 2025. 2 on PithPith's reviewarXiv
- [23] Xin Lin, Meixi Song, Dizhe Zhang, Wenxuan Lu, Haodong Li, Bo Du, Ming-Hsuan Yang, Truong Nguyen, and Lu Qi. Depth any panoramas: A foundation model for panoramic depth estimation. InProceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 2683... on Pith
- [24] Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization.arXiv preprint arXiv:1711.05101, 2017. 5 on PithPith's reviewarXiv
- [25] Kai Luo, Hao Shi, Sheng Wu, Fei Teng, Mengfei Duan, Chang Huang, Yuhang Wang, Kaiwei Wang, and Kailun Yang. Omnidirectional multi-object tracking. In2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 21959–21969. IEEE, 2025. 1 on Pith
- [26] Mapillary. Metropolis Dataset, 2024. Accessed: 2025-11- on Pith
- [27] Chao Ning and Hongping Gan. Trap attention: Monocu- lar depth estimation with manual traps. In2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 5033–5043. IEEE, 2023. 4 on Pith
- [28] Maxime Oquab, Timoth ´ee Darcet, Th ´eo Moutakanni, Huy V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023. 1 on PithPith's reviewarXiv
- [29] Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas K ¨opf, Edward Yang, Zach DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu F... on Pith
- [30] Gr ´egoire Payen de La Garanderie, Amir At- apour Abarghouei, and Toby P Breckon. Eliminating the blind spot: adapting 3d object detection and monocular depth estimation to360 ◦ panoramic imagery. InEuropean Conference on Computer Vision, pages 812–830. Springer, on Pith
- [31] Ren ´e Ranftl, Katrin Lasinger, David Hafner, Konrad Schindler, and Vladlen Koltun. Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer.IEEE transactions on pattern analysis and machine intelligence, 44(3):1623–1637, 2020. 2 on Pith
- [32] Ren ´e Ranftl, Alexey Bochkovskiy, and Vladlen Koltun. Vi- sion transformers for dense prediction. In2021 IEEE/CVF International Conference on Computer Vision (ICCV), pages 12159–12168. IEEE, 2021. 2 on Pith
- [33] Zhijie Shen, Chunyu Lin, Kang Liao, Lang Nie, Zishuo Zheng, and Yao Zhao. Panoformer: panorama transformer for indoor360 ◦ depth estimation. InEuropean Conference on Computer Vision, pages 195–211. Springer, 2022. 2 on Pith
- [34] Oriane Sim ´eoni, Huy V V o, Maximilian Seitzer, Federico Baldassarre, Maxime Oquab, Cijo Jose, Vasil Khalidov, Marc Szafraniec, Seungeun Yi, Micha ¨el Ramamonjisoa, et al. Dinov3.arXiv preprint arXiv:2508.10104, 2025. 1 on PithPith's reviewarXiv
- [35] Leslie N Smith and Nicholay Topin. Super-convergence: Very fast training of neural networks using large learn- ing rates. InArtificial intelligence and machine learning for multi-domain operations applications, pages 369–386. SPIE, 2019. 5 on Pith
- [36] Chuanxin Song and Peixi Peng. Pvdepth: Panoramic video depth estimation via geometry-aware spatiotemporal adap- tation. InForty-third International Conference on Machine Learning. 2, 3, 9 on Pith
- [37] Tencent. HunyuanWorld 1.0: Generating immersive, explorable, and interactive 3D worlds from words or pixels.https : / / 3d . hunyuan . tencent . com / sceneTo3D, 2025. 2 on Pith
- [38] Fei Teng, Kai Luo, Sheng Wu, Siyu Li, Pujun Guo, Jiale Wei, Jiaming Zhang, Kunyu Peng, and Kailun Yang. Hallucinat- ing 360: Panoramic street-view generation via local scenes diffusion and probabilistic prompting. In2026 IEEE Inter- national Conference on Robotics and Automati... on Pith
- [39] Fabio Tosi, Luca Bartolomei, Matteo Poggi, and Stefano Mattoccia. Zipdepth: Bringing lightweight zero-shot monoc- ular depth anywhere, on any device. InEuropean Conference on Computer Vision, pages 668–690. Springer, 2026. 3 on Pith
- [40] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need.Advances in neural information processing systems, 30, 2017. 2 on Pith
- [41] Matthew Wallingford, Anand Bhattad, Aditya Kusupati, Vivek Ramanujan, Matt Deitke, Sham Kakade, Anirud- dha Kembhavi, Roozbeh Mottaghi, Wei-Chiu Ma, and Ali Farhadi. From an image to a scene: Learning to imagine the world from a million 360 videos.Advances in Neural Informatio... on Pith
- [42] Fu-En Wang, Yu-Hsuan Yeh, Min Sun, Wei-Chen Chiu, and Yi-Hsuan Tsai. Bifuse: Monocular 360 depth estimation via bi-projection fusion. In2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 459–468. IEEE, 2020. 1, 2 on Pith
- [43] Fu-En Wang, Yu-Hsuan Yeh, Yi-Hsuan Tsai, Wei-Chen Chiu, and Min Sun. Bifuse++: Self-supervised and efficient bi-projection fusion for 360 depth estimation.IEEE trans- actions on pattern analysis and machine intelligence, 45(5): 5448–5460, 2022. 2, 6 on Pith
- [44] Guangcong Wang, Peng Wang, Zhaoxi Chen, Wenping Wang, Chen Change Loy, and Ziwei Liu. Perf: Panoramic neural radiance field from a single panorama.IEEE Transac- tions on Pattern Analysis and Machine Intelligence, 46(10): 6905–6918, 2024. 1 on Pith
- [45] Ning-Hsu Wang and Yu-Lun Liu. Depth anywhere: Enhanc- ing 360 monocular depth estimation via perspective distilla- tion and unlabeled data augmentation.Advances in Neural Information Processing Systems, 37:127739–127764, 2024. 2, 3 on Pith
- [46] Pengfei Wang, Shihao Wang, Liyi Chen, Zhiyuan Ma, Guowen Zhang, and Lei Zhang. Depthmaster: Unified monocular depth estimation for perspective and panoramic images.arXiv preprint arXiv:2606.12368, 2026. 1, 2, 6 on PithPith's reviewarXiv
- [47] Ruicheng Wang, Sicheng Xu, Yue Dong, Yu Deng, Jianfeng Xiang, Zelong Lv, Guangzhong Sun, Xin Tong, and Jiaolong Yang. Moge-2: Accurate monocular geometry with metric scale and sharp details.Advances in Neural Information Pro- cessing Systems, 38:35928–35959, 2026. 2, 3 on Pith
- [48] Wenshan Wang, Delong Zhu, Xiangwei Wang, Yaoyu Hu, Yuheng Qiu, Chen Wang, Yafei Hu, Ashish Kapoor, and Se- bastian Scherer. Tartanair: A dataset to push the limits of visual slam. In2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 4909–491... on Pith
- [49] Diana Wofk, Fangchang Ma, Tien-Ju Yang, Sertac Karaman, and Vivienne Sze. Fastdepth: Fast monocular depth esti- mation on embedded systems. In2019 International Confer- ence on Robotics and Automation (ICRA), pages 6101–6108. IEEE, 2019. 3 on Pith
- [50] Yinzhe Xu, Huajian Huang, Yingshu Chen, and Sai-Kit Ye- ung. 360vots: Visual object tracking and segmentation in omnidirectional videos.IEEE Transactions on Pattern Anal- ysis and Machine Intelligence, 47(11):9785–9797, 2025. 1 on Pith
- [51] Feng Xue, Wu Chen, Mingshuai Zhao, Guofeng Zhong, Anlong Ming, Haozhe Wang, Dianqiao Lei, Zhaowen Lin, Haiyang Zhang, and Nicu Sebe. Depthart: Scaling foun- dation monocular depth to tiny models.arXiv preprint arXiv:2607.17099, 2026. 8 on PithPith's reviewarXiv
- [52] Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, and Hengshuang Zhao. Depth anything: Unleashing the power of large-scale unlabeled data. In2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 10371–10381. IEEE, 2024. 2 on Pith
- [53] Lihe Yang, Bingyi Kang, Zilong Huang, Zhen Zhao, Xiao- gang Xu, Jiashi Feng, and Hengshuang Zhao. Depth any- thing v2.Advances in neural information processing sys- tems, 37:21875–21911, 2024. 2 on Pith
- [54] Ilwi Yun, Chanyong Shin, Hyunku Lee, Hyuk-Jae Lee, and Chae Eun Rhee. Egformer: Equirectangular geometry- biased transformer for 360 depth estimation. In2023 IEEE/CVF International Conference on Computer Vision (ICCV), pages 6078–6089. IEEE, 2023. 2 on Pith
- [55] Jiaming Zhang, Kailun Yang, Chaoxiang Ma, Simon Reiß, Kunyu Peng, and Rainer Stiefelhagen. Bending reality: Distortion-aware transformers for adapting to panoramic se- mantic segmentation. In2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 1689... on Pith
- [56] Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. In 2023 IEEE/CVF International Conference on Computer Vi- sion (ICCV), pages 3813–3824. IEEE, 2023. 5 on Pith
- [57] Ning Zhang, Francesco Nex, George V osselman, and Nor- man Kerle. Lite-mono: A lightweight cnn and transformer architecture for self-supervised monocular depth estimation. In2023 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 18537–18546. IEEE, 2023. 4 on Pith
- [58] Weiming Zhang, Yexin Liu, Xu Zheng, and Lin Wang. Goodsam: Bridging domain and capacity gaps via seg- ment anything model for distortion-aware panoramic seman- tic segmentation.arXiv preprint arXiv:2403.16370, 2024. 1 on PithPith's reviewarXiv
- [59] Jia Zheng, Junfei Zhang, Jing Li, Rui Tang, Shenghua Gao, and Zihan Zhou. Structured3d: A large photo-realistic dataset for structured 3d modeling. InEuropean Conference on Computer Vision, pages 519–535. Springer, 2020. 2, 3 on Pith
- [60] Ding Zhong, Xu Zheng, Chenfei Liao, Yuanhuiyi Lyu, Jialei Chen, Shengyang Wu, Linfeng Zhang, and Xuming Hu. Om- nisam: Omnidirectional segment anything model for uda in panoramic semantic segmentation. In2025 IEEE/CVF In- ternational Conference on Computer Vision (ICCV), pages... on Pith
- [61] Yichao Zhou, Jingwei Huang, Xili Dai, Shichen Liu, Linjie Luo, Zhili Chen, and Yi Ma. Holicity: A city-scale data platform for learning holistic 3d structures.arXiv preprint arXiv:2008.03286, 2020. 2, 3 on PithPith's reviewarXiv
- [62] Chuanqing Zhuang, Zhengda Lu, Yiqun Wang, Jun Xiao, and Ying Wang. Acdnet: Adaptively combined dilated con- volution for monocular panorama depth estimation. InPro- ceedings of the AAAI conference on artificial intelligence, pages 3653–3661, 2022. 1, 2, 6, 8 on Pith
- [63] Nikolaos Zioulis, Antonis Karakottas, Dimitrios Zarpalas, and Petros Daras. Omnidepth: Dense depth estimation for indoors spherical panoramas. InEuropean conference on computer vision, pages 453–471. Springer, 2018. 2 on Pith
Challenges and discussion
Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.
Come back by email
A short list of the week's newly reviewed papers. Nothing is sent until you confirm from your inbox, and every email has a link that stops it. No account.
No comments yet.
Sign in to comment