The DOI matches its Crossref record; 57 of the 59 references it cites were found in a registry or an index; none of the 18 sources its claims cite is checked for support yet.
3D Packing for Self-Supervised Monocular Depth Estimation
Open accessAbstract
Although cameras are ubiquitous, robotic platforms typically rely on active sensors like LiDAR for direct 3D perception. In this work, we propose a novel self-supervised monocular depth estimation method combining geometry with a new deep network, PackNet, learned only from unlabeled monocular videos. Our architecture leverages novel symmetrical packing and unpacking blocks to jointly learn to compress and decompress detail-preserving representations using 3D convolutions. Although self-supervised, our method outperforms other self, semi, and fully supervised methods on the KITTI benchmark. The 3D inductive bias in PackNet enables it to scale with input resolution and number of parameters without overfitting, generalizing better on out-of-domain data such as the NuScenes dataset. Furthermore, it does not require large-scale supervised pretraining on ImageNet and can run in real-time. Finally, we release DDAD (Dense Depth for Automated Driving), a new urban driving dataset with more challenging and accurate depth evaluation, thanks to longer-range and denser ground-truth depth generated from high-density LiDARs mounted on a fleet of self-driving cars operating world-wide.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
PackNet is a self-supervised monocular depth architecture using symmetrical 3D packing/unpacking blocks, trained on unlabeled monocular videos, optionally with velocity supervision for metric scale. It reports state-of-the-art KITTI results, better generalization on NuScenes, real-time inference, and releases the DDAD dataset for long-range dense depth evaluation.
Key findings
- On DDAD, PackNet-SfM achieves Abs Rel 0.162, Sq Rel 3.917, RMSE 13.452, RMSE log 0.269, and δ<1.25 0.823, outperforming Monodepth2 (R18) at Abs Rel 0.381 and RMSE 21.277; the gap increases at longer ranges up to 200m.
- On NuScenes generalization without fine-tuning, PackNet reaches Abs Rel 0.187, Sq Rel 1.852, RMSE 7.636, RMSE log 0.289, and δ<1.25 0.742, better than ResNet50 (Abs Rel 0.216, RMSE 8.477) and ResNet50‡ (Abs Rel 0.210, RMSE 8.111).
- Ablation on KITTI at 640×192 shows PackNet with D=8 3D filters reaches Abs Rel 0.111, Sq Rel 0.785, RMSE 4.601, RMSE log 0.189, and δ<1.25 0.878, improving over no pack/unpack (Abs Rel 0.121, RMSE 4.831).
- PackNet reaches about 128M parameters with 60ms inference on a Titan V100, improvable to <30ms with TensorRT; a packing/unpacking reconstruction test achieves loss 0.0079 versus 0.063 for max-pooling/bilinear upsampling.
Methods
- Self-supervised monocular SfM jointly trains a depth network and pose network on unlabeled monocular videos using a photometric loss (SSIM + L1), minimum over source images, auto-masking, and edge-aware depth smoothness; an optional velocity supervision loss aligns pose-translation magnitude with measured velocity to resolve scale ambiguity.
- PackNet is a symmetrical encoder-decoder with packing blocks (Space2Depth, 3D convolution expansion, 2D contraction) and unpacking blocks (2D convolution, 3D convolution, Depth2Space), plus skip connections and multiple inverse-depth outputs.
- Training uses KITTI (39,810 train, 4,424 validation, 697 evaluation; improved ground truth 652 maps), DDAD (17,050 train and 4,150 evaluation frames, Luminar-H2, up to 200m), NuScenes (6,019 front-facing validation images), and CityScapes (88,250 training images) for pretraining; PyTorch on 8 Titan V100 GPUs, Adam, 100 epochs, batch size 4, depth LR 2e-4, pose LR 5e-4, α=0.85, λ1=0.001, λ2=0.05.
Limitations
- The paper states monocular SfM is inherently scale-ambiguous; metric scale requires velocity supervision at training time, otherwise test-time LiDAR ground-truth scaling is needed.
- The paper states DDAD experiments use only the front-facing camera, and that packing/unpacking do not ensure invertibility.
- No explicit limitations section is provided; evident constraints include KITTI evaluation up to 80m and DDAD up to 200m, and pose results are deferred to supplementary material due to space.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
27 Own result · 14 Not checked yet · 13 No basis given · 12 Own interpretation
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“iii) generalizes better on unseen data;”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 1. Introduction
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“ii) is competitive with fully supervised methods;”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 1. Introduction
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“While recent works in self-supervised monocular depth estimation have mostly focused on engineering the loss function [6,34,48,54], we show that performance critically depends on the model architecture, in line with the observations of [28] for other self-supervised tasks.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 1. Introduction
What the sources say
- Not checked yetcites [28] source
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“v) can run in real time at high resolution;”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 1. Introduction
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our experiments on the standard KITTI benchmark [17], the recent NuScenes dataset [5], and our new proposed DDAD benchmark show that our self-supervised monocular approach i) improves on the state of the art, especially at longer ranges;”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 1. Introduction
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“vi) does not require supervised pretraining on ImageNet to achieve state-of-the-art results;”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 1. Introduction
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“or test-time ground-truth scaling if velocity information is available at training time.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 1. Introduction
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
- Rests on an assumption: Instantaneous velocity measurements are ubiquitous in most mobile systems today.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, we release DDAD (Dense Depth for Automated Driving), a new urban driving dataset with more challenging and accurate depth evaluation, thanks to longer-range and denser ground-truth depth generated from high-density LiDARs mounted on a fleet of self-driving cars operating world-wide.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“especially as more 3D convolutional filters are added”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.6. Ablation Studies
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Furthermore, we show that by simply introducing an additional source of unlabeled videos, such as the publicly available CityScapes dataset (CS+K) [9], we are able to further improve monocular depth estimation performance.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.3. Depth Estimation Performance
What the sources say
- Not checked yetcites [9] source
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- It may state two separate claims as one. 17
- The quote only partly states this claim. 1
- Probably not a substantial claim. 7
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“iv) scales better with number of parameters, input resolution, and more unlabeled training data;”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 1. Introduction
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Additionally, by leveraging during training only weak velocity information, we are able to make our model scale-aware, i.e. producing metrically accurate depth maps from a single image.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 6. Conclusion
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Although purely trained on unlabeled monocular videos, our approach outperforms other existing selfand semi-supervised methods and is even competitive with fully-supervised methods while able to run in real-time.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 6. Conclusion
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“As indicated by Pillai et al. [39], we also observe an improvement in performance at higher image resolutions, which we attribute to the proposed network's ability to properly preserve and process spatial information end-to-end.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.3. Depth Estimation Performance
What the sources say
- Not checked yetcites [39]
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This new dataset is a more realistic and challenging benchmark for depth estimation, as it is diverse and captures precise structure across images (30k points per frame) at longer ranges (up to 200m vs 80m for previous datasets).”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > DDAD (Dense Depth for Automated Driving).
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“and further improves as more complexity is added”
- Also stated in: § Input image · boosted
“improves significantly with more parameters in both cases without overfitting.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.5. Network Complexity
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Also stated in: § Input image · boosted
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The ATE results are summarized in Table 6, with our proposed framework achieving competitive results relative to other related methods.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > A. Pose evaluation
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one. The quote only partly states this claim.
In the paper
“We use the architecture proposed by [53] without the explainability mask, which we found not to improve results.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > Depth Network.
What the sources say
- Not checked yetcites [53] source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It includes a highresolution, long-range Luminar-H2foot_0 as the LiDAR sensor used to generate pointclouds, with a maximum range of 250m and sub-1cm range precision.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > B. Dense Depth for Automated Driving (DDAD)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It contains 17,050 training and 4,150 evaluation frames with ground-truth depth maps generated from dense LiDAR measurements using the Luminar-H2 sensor.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > DDAD (Dense Depth for Automated Driving).
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“and can be further improved by self-supervised pretraining on other datasets”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.6. Ablation Studies
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
None of the sources cited for this claim have been checked yet.
In the paper
“Our best results are achieved when injecting both more unlabeled data at training time and processing higher resolution input images, achieving performance comparable to semisupervised [29] and fully supervised [15] methods.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.3. Depth Estimation Performance
What the sources say
- Not checked yetcites [29]
- Not checked yetcites [15]
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the proposed architecture (Table 1) reaches around 128M parameters with an inference time of 60ms on a Titan V100 GPU”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.5. Network Complexity
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“From these results we can see that our PackNet-SfM approach significantly outperforms the state-of-the-art [19], based on the ResNet family, the performance gap consistently increasing when larger distances are considered.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.3. Depth Estimation Performance
What the sources say
- Not checked yetcites [19]
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“DDAD is a cross-continental dataset with scenes drawn from urban settings in the United States (San Francisco Bay Area, Detroit and Ann Arbor) and Japan (Tokyo and Odaiba).”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > B. Dense Depth for Automated Driving (DDAD)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Additionally, it contains six calibrated cameras time-synchronized at 10 Hz, that together produce a 360 • coverage around the vehicle.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > B. Dense Depth for Automated Driving (DDAD)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Note that in the case of monocular SfM both depth and pose are estimated up to an unknown scale factor, due to the inherent ambiguity of the photometric loss.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 3. Self-Supervised Scale-Aware SfM
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This additional velocity loss allows the pose network to make metrically accurate predictions, subsequently resulting in the depth network also learning metrically accurate estimates to maintain consistency (cf.Section 5.4).”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > Velocity Supervision Loss.
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Rests on an assumption: Instantaneous velocity measurements are ubiquitous in most mobile systems today.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“as evidence that it does not simply memorize training data but learns transferable discriminative features.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.7. Generalization Capability
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The 3D inductive bias in PackNet enables it to scale with input resolution and number of parameters without overfitting, generalizing better on out-of-domain data such as the NuScenes dataset.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > Abstract
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“As mentioned in [15,19], ResNet architectures highly benefit from ImageNet pretraining, since they were originally developed for classification tasks.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.6. Ablation Studies
What the sources say
- Not checked yetcites [15]
- Not checked yetcites [19]
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The paper relies on this without support.
Instantaneous velocity measurements are ubiquitous in most mobile systems today.
“Since instantaneous velocity measurements are ubiquitous in most mobile systems today, we show that they can be directly incorporated in our self-supervised objective to learn a metrically accurate and scale-aware monocular depth estimator.”
The paper relies on this without support.
The same pixel will not be occluded or out-of-bounds in all context images, and the association with minimal photometric loss is correct.
“The intuition is that the same pixel will not be occluded or out-of-bounds in all context images, and that the association with minimal photometric loss should be the correct one.”
The paper relies on this without support.
Depth maps made by projecting LiDAR pointclouds onto the camera frame are accurate enough to serve as ground truth for evaluating monocular depth estimation.
“The depth maps generated from projecting these Luminar pointclouds onto the camera frame allow us to evaluate depth estimation methods in a much more challenging way, both in terms of denseness and longer ranges.”
“At test-time, all monocular methods (M) scale estimated depths with median ground-truth LiDAR information.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“some methods incorporate more 3D computations into our object detection pipeline by applying a 3D reconstruction method like [4, 5, 6] to create a pseudo-LiDAR or range input of the scene from camera images.”
From DETR3D: 3D Object Detection from Multi-view Images via 3D-to-2D Queries (2021) · cites this paper as [6]
In the citing paper
“The dataset provides semantic segmentation labels only for the validation and test (non-public) sets, significantly restricting its use-case.”
From Multimodal semantic segmentation in autonomous driving: A review of current approaches and future perspectives (2022) · cites this paper as [82]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 3D Packing for Self-Supervised Monocular Depth Estimation > 5.2. Implementation Details
“We use PyTorch [38] with all models trained across 8 Titan V100 GPUs.”
§ 3D Packing for Self-Supervised Monocular Depth Estimation > References
“38. Jeff Michels, Ashutosh Saxena, and Andrew Y Ng. High speed obstacle avoidance using monocular vision and rein- forcement learning.”
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
Image quality assessment: from error visibility to structural similarity
Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004) · IEEE Transactions on Image Processing · doi:10.1109/tip.2003.819861 · Cited by 57,892
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
(2014) · Cited by 33,419
The Cityscapes Dataset for Semantic Urban Scene Understanding
Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., & Schiele, B. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.350 · Cited by 12,433
(2017) · Cited by 11,625
Image Super-Resolution Using Deep Convolutional Networks
Dong, C., Loy, C. C., He, K., & Tang, X. (2016) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2015.2439281 · Cited by 10,105
Vision meets robotics: The KITTI dataset
Geiger, A., Lenz, P., Stiller, C., & Urtasun, R. (2013) · The International Journal of Robotics Research · doi:10.1177/0278364913491297 · Cited by 9,913
Research works printed but not linked: 2
- kingma, 0, Glow: Generative flow with invertible 1×1 convolutions, Advances in neural information processing systems
- graham, 2015, Fractional Max-Pooling
Cited by
Sorted by citations
Items 701–739 of 739
On depth prediction for autonomous driving using self-supervised learning
Boulahbal, H. E. (2023) · theses.fr (ABES) · doi:10.70675/273dde48zd8b7z42b8z9c85z52bfc7aa4bba
Self-Supervised Monocular Depth Estimation With Multimodal Fusion and Pose Enhancement
Chen, H., Wang, Y., Jia, S., Huang, S., & Zhang, J. (2026) · IEEE Sensors Journal · doi:10.1109/jsen.2026.3684347
Instance-Level Video Depth in Groups Beyond Occlusions
Liang, Y., Zhou, Y., Sun, Z., Xiang, T., Li, G., & He, S. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00711
Too Tiny to See: Hazardous Obstacle Detection Dataset and Evaluation
Miekkala, T., Brucker, S., Walz, S., Ghilotti, F., Ramazzina, A., Scheuble, D., Pyykönen, P., Bijelic, M., & Heide, F. (2026) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv69130.2026.00184
UniLiPs: Unified LiDAR Pseudo-Labeling with Geometry-Grounded Dynamic Scene Decomposition
Ghilotti, F., Brucker, S., Saidy, N., Matteucci, M., Bijelic, M., & Heide, F. (2026) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv69130.2026.00160 · Cited by 1
SA4Depth: Consistent Pose-Depth Scale Alignment for Self-Supervised Monocular Depth Estimation
Li, C., Berner, N., Navab, N., Tombari, F., & Gasperini, S. (2026) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2026.3699248
Zhao, B., He, H., Wu, T., Zhao, G., Hao, X., & Huang, G. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3699753
Zhang, X., Li, J. F., Liu, C., Liu, C., Li, J. F., & Yang, D. (2026) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2026.3701560
Nakhon, R. N., Chanthakulkit, P., Sutthaweekul, R., Lee, W., Srisomboon, K., & Chen, X. (2025) · IEEE Far East NDT New Technology & Application Forum (FENDT) · doi:10.1109/fendt66689.2025.11547544
Bai, H., Li, S., Zhai, H., Gao, Y., Lv, C., Wang, Y., Ping, H., Hua, W., & Gao, X. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3703317
Xue, W., & Huo, Z. (2026) · International Conference on Multimedia Retrieval · doi:10.1145/3805622.3810646
Vigdepth: self-supervised monocular depth estimation based on vision GNN and local graph reasoning
Chen, X., & Wang, M. (2026) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-026-03185-3
Li, Z., Zhu, K., & Jin, G. (2026) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-026-02062-3
AerialMetric: Benchmarking and Adapting UAV Monocular Metric Depth Estimation in the Real World
Song, Z., Chen, G., Zhang, Y., Zou, Y., Fu, C., Yuan, Z., Huang, C., Cui, S. R., & Cao, X. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-36984-0_20
Zhu, Z. G., Zhang, L., Chen, Y., Zhang, T., Wang, Y., & Zhan, M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-36846-1_7
Koledić, K., Petrović, L., Petrović, I., & Marković, I. (2026) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2026.105617 · Cited by 1
Wang, A., Liao, G., Zhao, Z., Li, R., Fang, Z., Ma, S., & Hwang, J. (2026) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2026.3711298
FoundationGeo: Learning Spatial Pixel-Wise Fields for Monocular Metric Geometry
Liu, M., Lyu, X., Ren, T., Dai, P., Wu, X., Zhang, Z., Zhang, J., Lin, J., Shi, S., & Qi, X. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37092-1_20
Zhu, Z., Sun, J., & Han, P. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-3548-3_16
Semantic-Guided Self-Supervised Monocular Depth Estimation With Context-Aware Attention
Hu, Y., & Xiao, Y.-Y. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-3429-5_46
Geometric Distillation from Rectified Stereo: Leveraging Epipolar Cues for Monocular Depth
Kim, J.-H., & Liu, X. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37362-5_34
SteerCNN: An FPGA-Based Neural Network Control System for Autonomous Vehicle Steering
Mahambrey, A. A., & Ramesh, C. (2026) · International Conference on Computer Networks, Big Data and IoT (ICCBI) · doi:10.1109/iccbi68589.2026.11619684
Shanto, H. M., Islam, M. T., Hasan, M. R., Pranto, S. S., Ahmed, T., Mahmud, R., & Tasnim, S. A. (2026) · AIUB Journal of Science and Engineering (AJSE) · doi:10.53799/zsh31s95
MGD-Depth: Disentangling Scene Dynamics via Multi-granularity Representation Learning
Yue, S., Ren, Y., Li, J., Peng, K., & He, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31583-0_44
FeDepth: Federated Learning for Depth Estimation Under Robot Heterogeneity
Lee, G., Lee, I., Lee, J., Lee, J., Yoon, S. W., & Joo, K. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37574-2_17
Chen, Q., Wei, C., Zhu, B., Yan, Q., Chen, S., & Zhou, D. (2026) · Journal of Imaging · doi:10.3390/jimaging12080362
Nguyen, Q. H., & Albertella, A. (2026) · Geomatics · doi:10.3390/geomatics6050102
MetricAnything: Scaling Metric Depth Pretraining with Noisy Heterogeneous Sources
Yang, J., Di, D., Zhang, X., Fan, L., Cui, J., Li, H., & Ma, B. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37016-7_16 · Cited by 6
Liu, J., Lu, G., Wang, Y., Wei, L., & Chen, D. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.114891
Consistent Monocular Depth Estimation with Contact Region Boundary-Aware Refinement
Wang, Y., Miao, Q., & Zuo, W. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37624-4_10
Zhong, Y., Hao, Y., Wu, X., Liao, T., & Wang, C. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.114984
Song, X., Liu, J., Zhao, Q., Guo, X., Wei, B., & Hei, X. (2026) · The Visual Computer · doi:10.1007/s00371-026-04748-9
Range-Aware Camera-LiDAR Fusion for Efficient Long-Range Omnidirectional Depth Estimation
Seo, Y., & Kim, C. (2026) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2026.3739565
RAD: Retrieval-Augmented Monocular Metric Depth Estimation for Underrepresented Classes
Baltaxe, M., Levi, D., & Benaim, S. (2026) · doi:10.1109/cvprf71909.2026.00058
Towards Robust Scale Consistency in Self-supervised Monocular Video Depth Estimation
Do, T.-H., & Pham, H.-A. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-6513-8_11
Afxentiou, V., & Vladimirova, T. (2026) · IEEE Access · doi:10.1109/access.2026.3742560
Prompt Bins Learning for Monocular Depth Estimation
Wang, Y., Wang, Z., Li, J., Liu, Q., & Xu, J. (2026) · Chinese Journal of Electronics · doi:10.23919/cje.2025.00.496
JEPADepth: Self-Supervised Monocular Depth Estimation via Joint-Embedding Predictive Architectures
Grigore-Atimuț, I.-O., & Popa, C.-A. (2026) · Mathematics · doi:10.3390/math14213846
Yao, P., Xiang, L., Wang, Y., & Fan, B. (2026) · Intelligent Marine Technology and Systems · doi:10.1007/s44295-026-00120-8
Metadata
- Authors
- Guizilini, V., Ambrus, R., Pillai, S., Raventos, A., & Gaidon, A.
- Affiliations
- Toyota Research Institute (US)
- Venue
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2020-06
- Pages
- 2482-2491
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2991840778
- Semantic Scholar
- 43fcb6cf679cd2ae412d7cb53e85312600cc20dd
- Cited by
- 741
- References
- 59
- Citations by year
- 2026: 51 · 2025: 108 · 2024: 140 · 2023: 155 · 2022: 139 · 2021: 125 · 2020: 19
- FWCI
- 33.78
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Keywords
- monocular depth estimation, self-supervised learning, 3D convolution, out-of-domain generalization, real-time depth estimation, autonomous driving, KITTI dataset, nuScenes
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-08, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-08 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
